<?xml version="1.0" encoding="UTF-16"?><html xmlns="http://www.w3.org/1999/xhtml"><head><link rel="stylesheet" type="text/css" href="format.css" /></head><body bgcolor="#000000" text="#FFFFFF" link="#33CCFF" vlink="#33CCFF" alink="#FF0000"><a name="top" id="top" /><p><a href="http://syntheticspeech.de/index.html">site navigation</a></p><hr /><h1>German Text-to-Speech</h1><p>last update: <font color="#FF0000">24th April 2018</font></p><h3>Contents:</h3><ol><li><a href="#foreword">Foreword</a></li><li><a href="#commercial">Commercial systems</a></li><li><a href="#universities">Universities/research</a></li><li><a href="#others">Other systems</a></li><li><a href="#service">Service systems</a></li><li><a href="#other">Further samples</a></li><li><a href="#licensed">Licensed products</a></li><li><a href="#missing">Missing examples</a></li><li><a href="#unknown">Unknown examples</a></li><li><a href="#ttsChart">TTS classification chart</a></li><li><a href="#credits">Credits</a></li><li><a href="#changelog">Change-log</a></li></ol><hr /><p><a name="foreword" id="foreword" /></p><h3>Remarks</h3><ul><li>This collection of German TTS samples is maintained by <a href="http://felix.syntheticspeech.de/">Felix Burkhardt</a></li><li>There's also a page on <a href="http://emosamples.syntheticspeech.de/index.html">Speechsynthesis demos with simulated emotion</a></li><li>I appreciate hints about missing systems! ( <a href="mailto:felixbur@gmx.de" rev="made">felixbur@gmx.de</a> )</li><li>Generally the demonstrations don't show the up-to-date quality of the systems!</li><li>The information provided is to my best knowledge, but of course I can't guarantee for the correctness!</li><li>Some difference in quality is due to different sample-rates; most demos are 16 kHz but some are 8 kHz or 22.</li><li>Don't rely on the year specifications, I'm very unsure about most of them. Many of them simply denote the year when I took the samples (although they're meant to stand for the year when the voice / engine was released).</li><li>The remark "courtesy of company X" means that I got the samples specially for this page from the respective company. Possibly they made special adjustments wrt. pronounciation and prosody.</li></ul><h3>Terminology</h3><p>I added a <a href="#ttsChart">chart</a> to facilitate the understanding of the concepts used for classification. It's kind of out-dated, as non-uniform unit selection is not explicitely mentioned.</p><p>TTS consists always of two components, which I call Dutoit's introduction):</p><ul><li><b>NLP</b> (Natural Language Processing): conversion of orthographical text into phoneme-alphabet and prosody description.</li><li><b>DSP</b> (Digital Speech Processing): speech engine: synthesis of speechsignal from ouput of NLP-component.</li></ul><p>The engines that synthesize the speech (DSP-component) are based mainly on five main technologies:</p><ul><li><b>DNN Synthesis</b>: Quite the newest addition to speech synthesis algorithms are artificial neural networks (or deep neural nets, meaning the number of layers is higher than in traditional artificial neural network architecture, say: five. TTS with neural nets has been done since many decades but to my knowledge not for German). They replace the HMM approach to predict the best acoustic parameters for a given sequence of symbols representing text.</li><li><b>HMM Synthesis</b>: Synthesis based on Hidden Markov Models, a statistical approach to model the transition probabilites of the acoustic parameters based on the speech to be generated. The approaches are trained on a relatively large data corpus, but have a small footprint for synthesis because they don't operate on the wavedata directly but on some parameterized representation (e.g. LPC). However this is also the reason they tend to produce artifacts. Sample: Simple4All <a href="ttsSamples/simple4all_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></li><li><b>Non-uniform unit-selection</b>: Best fitting chunks of speech from large databases get concatenated, minimizing a double cost-function: best fit to neighbor unit and best fit to target prosody. Sounds most natural (similar to original speaker), but unflexible and large footprint. Sample: RealSpeak <a href="ttsSamples/realSpeak_steffi1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></li><li><b>Diphone-synthesis</b>: Speech concatenated from diphone-units (two-phone combinations), prosody-fitting done by signal-manipulation (depends on unit-coding). relatively small footprint but not very natural. Sample: Bell-Labs synthesis <a href="ttsSamples/bell_labs_demo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></li><li><b>Formant-synthesis</b>: Speech synthesized by physical models (formants are resonance frequencies in vocal-tract). Very flexible and smallest footprint, but very unnatural. Sample: Eloquence <a href="ttsSamples/eloqdemo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></li></ul><hr /><p>The test sentences were:</p><p><b>sentence 1:</b></p><p><font color="#FF6666"><b>An den Wochenenden bin ich jetzt immer nach Hause gefahren und habe Agnes besucht. Dabei war eigentlich immer sehr schnes Wetter gewesen.</b></font></p><p>As I found this sentence a bit too simple, I thought up another test sentence which contains a collection of known problems for the NLP module: (in some demos this sentence is truncuated due to provider's restriction on character number)</p><p><b>sentence 2:</b></p><p><font color="#FF6666"><b>Dr. A. Smithe von der NATO (und nicht vom CIA) versorgt z.B. - meines Wissens nach - die Heroin seit dem 15.3.00 tgl. mit 13,84 Gramm Heroin zu 1,04 DM das Gramm.</b></font></p><p>Speaking now 6 years after thinking up those sentences, more pressing problems for German speech synthesis used in services like email-reading arise from the pronounciation of english terms, e.g. the following sentence would not be pronounced correctly by most systems without tuning:</p><p><b>sentence 3:</b></p><p><font color="#FF6666"><b>Die Manpowerdiskussion wird gecancelt, du kannst das File vom Server downloaden.</b></font></p><h3>Commercial <a name="commercial" id="commercial" /></h3><table border="0"><tr class="tableheader"><th width="15%"><b>company/link</b></th><th width="20%"><b>description/engine name</b></th><th width="25%"><b>technology</b></th><th width="10%"><b>languages</b></th><th width="10%"><b>voice name</b></th><th width="5%">year (approx.)</th><th width="5%">s1</th><th width="5%">s2</th><th width="5%">s3</th></tr><tr class="tablecell1"><td valign="top" rowspan="11"><a id="acapela" name="acapela" href="http://www.acapela-group.com" xmlns="">Acapela</a><br xmlns="" /><br xmlns="" />Acapela was formed in December 2003 from a combination of three European companies specializing in vocal technologies, Babel Technologies (Belgium), Infovox (Sweden) and Elan Speech (France).</td><td valign="top" rowspan="8" xmlns="">Acapela HQ TTS<a id="" name="">Acapela HQ TTS</a></td><td valign="top" rowspan="8" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="8" xmlns="">DE, FR, NL, ES, SE, US, SA, CY, DN, FI, CA, GR, IE, NO, PL, PT, BR, RU, TR</td><td valign="top" xmlns="">Claudia<br /></td><td valign="top" xmlns="">2015</td><td valign="top" xmlns=""><a href="ttsSamples/acapela_claudia_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_claudia_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_claudia_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Claudia (Smile)<br /></td><td valign="top" xmlns="">2017</td><td valign="top" xmlns=""><a href="ttsSamples/acapela_claudia_smile_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_claudia_smile_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_claudia_smile_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Lea (Child)<br /></td><td valign="top" xmlns="">2013</td><td valign="top" xmlns=""><a href="ttsSamples/acapela_lea_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_lea_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_lea_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Jonas (child)<br /></td><td valign="top" xmlns="">2013</td><td valign="top" xmlns=""><a href="ttsSamples/acapela_jonas_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_jonas_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_jonas_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Andreas<br /></td><td valign="top" xmlns="">2011</td><td valign="top" xmlns=""><a href="ttsSamples/acapela_andreas_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_andreas_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_andreas_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Julia<br /></td><td valign="top" xmlns="">2009</td><td valign="top" xmlns=""><a href="ttsSamples/acapela_Julia_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_Julia_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela_Julia_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Klaus<br /></td><td valign="top" xmlns="">2006</td><td valign="top" xmlns=""><a href="ttsSamples/brightSpeech_klaus_16kHz_s1.mp3&quot;"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/brightSpeech_klaus_16kHz_s2.mp3&quot;"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/brightSpeech_klaus_16kHz_s3.mp3&quot;"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Sarah<br /></td><td valign="top" xmlns="">2003</td><td valign="top" xmlns=""><a href="ttsSamples/bright12-sarah1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/bright12-sarah2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/bright12-sarah3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" rowspan="2" xmlns="">Custom Voice<a id="" name="">Custom Voice</a></td><td valign="top" rowspan="2" xmlns="">non-uniform unit-selection, ANN</td><td valign="top" rowspan="2" xmlns="">DE</td><td valign="top" xmlns="">Felix DNN (page author): <br />Artificial neural net model adapted with 15 minutes data</td><td valign="top" xmlns="">2017</td><td valign="top" xmlns=""><a href="ttsSamples/acapelaFelixDNN_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapelaFelixDNN_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapelaFelixDNN_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Felix (page author): <br />Non-uniform unit-selection from 2 hours data</td><td valign="top" xmlns="">2017</td><td valign="top" xmlns=""><a href="ttsSamples/acapelaFelix_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapelaFelix_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapelaFelix_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="1" xmlns="">Greeting Bunny<a id="" name="">Greeting Bunny</a></td><td valign="top" rowspan="1" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="1" xmlns="">DE, US, FR, IT, ES, NL, SE, NO, DK, BE</td><td valign="top" xmlns="">Bunny<br /></td><td valign="top" xmlns="">2008</td><td valign="top" xmlns=""><a href="ttsSamples/acapela-bunny_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela-bunny_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/acapela-bunny_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2"><td valign="top" rowspan="1"><a id="aculab" name="aculab" href="http://www.aculab.com/" xmlns="">Aculab</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="1" xmlns="">Aculab<a id="aculab" name="aculab">Aculab</a></td><td valign="top" rowspan="1" xmlns="">diphone<br />diphone-concatenation with LPC coded units. LPC (linear predictive coding), originally a compression algorithm, useful for synthesis because based on a source/filter model of speech.</td><td valign="top" rowspan="1" xmlns="">DE, UK, US, FR, BR, IT, ES</td><td valign="top" xmlns="">Julia<br /></td><td valign="top" xmlns="">1998</td><td valign="top" xmlns=""><a href="ttsSamples/aculab_1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aculab_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td></tr><tr class="tablecell1"><td valign="top" rowspan="6"><a id="aristech" name="aristech" href="http://aristools.de/" xmlns="">Aristech</a><br xmlns="" /><br xmlns="" />Formerly Speechconcept</td><td valign="top" rowspan="6" xmlns="">Cerevoice<a id="" name="">Cerevoice</a></td><td valign="top" rowspan="6" xmlns="">non-uniform unit-selection<br />Developments from Aristech, CereProc and University of Edinburgh</td><td valign="top" rowspan="6" xmlns="">DE, EN, FR, IT, ES, US, NL, JP</td><td valign="top" xmlns="">Sophie, adult<br />Corporate Voice, courtesy of Aristech</td><td valign="top" xmlns="">2011</td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s1-Sophie.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s2-Sophie.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s3-Sophie.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Leopold, Austrian adult<br />courtesy of Aristech</td><td valign="top" xmlns="">2013</td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s1-Leopold.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s2-Leopold.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s3-Leopold.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Alex, adult<br />courtesy of Aristech</td><td valign="top" xmlns="">2016</td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s1-Alex.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s2-Alex.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s3-Alex.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Gudrun, adult<br />courtesy of Aristech</td><td valign="top" xmlns="">2013</td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s1-Gudrun.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s2-Gudrun.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s3-Gudrun.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Nick, youth<br />courtesy of Aristech</td><td valign="top" xmlns="">2011</td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s1-Niclas.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s2-Niclas.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s3-Niclas.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Saskia, youth<br />courtesy of Aristech</td><td valign="top" xmlns="">2011</td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s1-Saskia.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s2-Saskia.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/aristech_s3-Saskia.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2"><td valign="top" rowspan="5"><a id="atip" name="atip" href="http://www.atip.de/" xmlns="">Atip</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="5" xmlns="">Proser<a id="" name="">Proser</a></td><td valign="top" rowspan="5" xmlns="">diphone<br />NLP-component and voices from Atip, Mbrola Engine (diphone-concatenation) from Babeltech</td><td valign="top" rowspan="5" xmlns="">DE, US</td><td valign="top" xmlns="">Carla<br /></td><td valign="top" xmlns="">2000</td><td valign="top" xmlns=""><a href="ttsSamples/proser_carla_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/proser_carla_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/proser_carla_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" xmlns="">Erkan<br />Turkish accent</td><td valign="top" xmlns="">2004</td><td valign="top" xmlns=""><a href="ttsSamples/proser_erkan_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/proser_erkan_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/proser_erkan_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Fifi <br />French accent</td><td valign="top" xmlns="">2004</td><td valign="top" xmlns=""><a href="ttsSamples/proser_fifi_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/proser_fifi_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/proser_fifi_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Steffen<br /></td><td valign="top" xmlns="">2000</td><td valign="top" xmlns=""><a href="ttsSamples/proser_steffen_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/proser_steffen_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/proser_steffen_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Eva<br /></td><td valign="top" xmlns="">2000</td><td valign="top" xmlns=""><a href="ttsSamples/proser_eva_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/proser_eva_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/proser_eva_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="2"><a id="att" name="att" href="http://www.digitalfuturesoft.com/attnaturalvoices.php" xmlns="">AT&amp;T</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="2" xmlns="">Natural Voices<a id="naturalvoices" name="naturalvoices">Natural Voices</a></td><td valign="top" rowspan="2" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="2" xmlns="">DE, IT, US, UK, FR, MX*</td><td valign="top" xmlns="">Klara<br /></td><td valign="top" xmlns="">2001</td><td valign="top" xmlns=""><a href="ttsSamples/att_klara_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/att_klara_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/att_klara_16kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Reiner<br /></td><td valign="top" xmlns="">2002</td><td valign="top" xmlns=""><a href="ttsSamples/att_reiner_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/att_reiner_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/att_reiner_16kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2"><td valign="top" rowspan="6"><a id="babeltech" name="babeltech" xmlns="">Babeltech</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="1" xmlns="">Brightspeech<a id="" name="">Brightspeech</a></td><td valign="top" rowspan="1" xmlns="">non-uniform unit-selection<br />same as Acapela HQ TTS</td><td valign="top" rowspan="1" xmlns=""></td><td valign="top" xmlns="">Ingrid<br /></td><td valign="top" xmlns="">2002</td><td valign="top" xmlns=""><a href="ttsSamples/brightSpeech_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/brightSpeech_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td><tr class="tablecell2"><td valign="top" rowspan="5" xmlns="">Babil<a id="" name="">Babil</a></td><td valign="top" rowspan="5" xmlns="">diphone<br />diphone-concatenation based on commercial Mbrola-engine. MBROLA (Multi Band Resynthesis Overlap and Add), similar to PSOLA but the database is treated beforehand to adapt pitch, amplitude and spectral features.</td><td valign="top" rowspan="5" xmlns="">DE, US, UK, ES, FR, NL, BE, BR, PT, IT, SE, NO, DK, FI, IS, TR, CZ, SA</td><td valign="top" xmlns="">Eva<br /></td><td valign="top" xmlns="">2000</td><td valign="top" xmlns=""><a href="ttsSamples/babil_eva_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/babil_eva_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/babil_eva_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" xmlns="">Greta<br /></td><td valign="top" xmlns="">2000</td><td valign="top" xmlns=""><a href="ttsSamples/babil_greta_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/babil_greta_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/babil_greta_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Steffen<br /></td><td valign="top" xmlns="">1997</td><td valign="top" xmlns=""><a href="ttsSamples/babil_steffen_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/babil_steffen_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/babil_steffen_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Helga<br />Same as Infovox 330</td><td valign="top" xmlns="">1998</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td></tr><tr class="tablecell2"><td valign="top" xmlns="">Gerhard<br />Same as Infovox 330</td><td valign="top" xmlns="">1998</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td></tr></tr></tr><tr class="tablecell1"><td valign="top" rowspan="1"><a id="belllabs" name="belllabs" xmlns="">Bell Labs</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="1" xmlns=""><a id="" name=""></a></td><td valign="top" rowspan="1" xmlns="">diphone<br />LPC-diphone concatenation</td><td valign="top" rowspan="1" xmlns="">DE, FR, ES, US, UK, IT, RU, RO, CN</td><td valign="top" xmlns=""><br /></td><td valign="top" xmlns=""></td><td valign="top" xmlns=""><a href="ttsSamples/bell_labs_demo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/bellLabs_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td></tr><tr class="tablecell2"><td valign="top" rowspan="1"><a id="centigram" name="centigram" xmlns="">Centigram</a><br xmlns="" /><br xmlns="" />Acquired by Lernout &amp; Hauspie, later Nuance</td><td valign="top" rowspan="1" xmlns="">TruVoice<a id="" name="">TruVoice</a></td><td valign="top" rowspan="1" xmlns="">formant</td><td valign="top" rowspan="1" xmlns="">DE, US, MX*, FR, IT</td><td valign="top" xmlns=""><br /></td><td valign="top" xmlns="">1996</td><td valign="top" xmlns=""><a href="ttsSamples/lh_formant.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/truVoice_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td></tr><tr class="tablecell1"><td valign="top" rowspan="2"><a id="cepstral" name="cepstral" href="http://www.cepstral.com/" xmlns="">Cepstral</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="2" xmlns="">Cepstral TTS<a id="" name="">Cepstral TTS</a></td><td valign="top" rowspan="2" xmlns="">non-uniform unit-selection<br />Associated wiith Alan Black, one of the pioneers of non-uniform unit-selection and lead scientist of Festival, an open source text-to-speech framework developed at Univ. of Edinburgh and the CMU.</td><td valign="top" rowspan="2" xmlns="">DE, UK, US, ES, FR, EG, TH, AF</td><td valign="top" xmlns="">Kathrin<br /></td><td valign="top" xmlns="">2003</td><td valign="top" xmlns=""><a href="ttsSamples/cepstral-katrin16-1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/cepstral-katrin16-2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/cepstral-katrin_16kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Matthias<br /></td><td valign="top" xmlns="">2003</td><td valign="top" xmlns=""><a href="ttsSamples/cepstral-matthias16-1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/cepstral-matthias16-2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/cepstral-matthias_16kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2"><td valign="top" rowspan="2"><a id="dtag" name="dtag" xmlns="">Deutsche Telekom</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="1" xmlns="">Berkom TTS<a id="" name="">Berkom TTS</a></td><td valign="top" rowspan="1" xmlns="">formant<br />Research system by former rd department of German Telekom. Hybrid approach combining formant synthesis for voiced phonemes and concatenating with waveform coded units for unvoiced parts.</td><td valign="top" rowspan="1" xmlns="">DE</td><td valign="top" xmlns="">Felix<br /></td><td valign="top" xmlns="">1998</td><td valign="top" xmlns=""><a href="ttsSamples/berkom_felix_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/berkom_felix_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/berkom_felix_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" rowspan="1" xmlns="">SAMT<a id="" name="">SAMT</a></td><td valign="top" rowspan="1" xmlns="">hardware-based formant synthesis<br />(Sprach-Ausgabesystem in Multiplex-Technik): hardware-based formant synthesis of former Forschungsinstitut der Deutschen Bundespost.</td><td valign="top" rowspan="1" xmlns="">DE</td><td valign="top" xmlns=""><br /><br /> Other sample: <a href="ttsSamples/samt.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">1987</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="1"><a id="dec" name="dec" xmlns="">Digital Equipment Corporation</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="1" xmlns="">DecTalk<a id="" name="">DecTalk</a></td><td valign="top" rowspan="1" xmlns="">formant<br />First commercial text-to-speech synthesizer. Rule based formant-synthesis - the legendary formant synthesizer, based on Klatt's MITTalk)</td><td valign="top" rowspan="1" xmlns="">DE, US, UK, ES, MX*, FR</td><td valign="top" xmlns=""><br /></td><td valign="top" xmlns="">1982</td><td valign="top" xmlns=""><a href="ttsSamples/decTalk_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/decTalk_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td></tr><tr class="tablecell2"><td valign="top" rowspan="3"><a id="elan" name="elan" xmlns="">Elan</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="1" xmlns="">SaySo<a id="" name="">SaySo</a></td><td valign="top" rowspan="1" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="1" xmlns="">DE, US, FR, IT, ES</td><td valign="top" xmlns="">Lea<br /></td><td valign="top" xmlns="">2003</td><td valign="top" xmlns=""><a href="ttsSamples/elanSaySo_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/elanSaySo_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/elanSaySo_lea_22kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" rowspan="2" xmlns="">Tempo<a id="" name="">Tempo</a></td><td valign="top" rowspan="2" xmlns="">diphone<br />Pitch Synchronous Overlap and Add (PSOLA): famous algorithm to change pitch and time of speech that made diphone-synthesis a great success for many years.</td><td valign="top" rowspan="2" xmlns="">DE, US, UK, FR, ES, IT, BR, PT, RU, PL</td><td valign="top" xmlns="">Thomas<br /></td><td valign="top" xmlns="">1998</td><td valign="top" xmlns=""><a href="ttsSamples/elan_thomas_1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/elan_thomas_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/elan_thomas_3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" xmlns="">Dagmar<br /></td><td valign="top" xmlns="">1996</td><td valign="top" xmlns=""><a href="ttsSamples/elan_dagmar_1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/elan_dagmar_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/elan_dagmar_3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr></tr><tr class="tablecell1"><td valign="top" rowspan="1"><a id="eloquent" name="eloquent" xmlns="">Eloquent Technologies</a><br xmlns="" /><br xmlns="" />Aquired by Scansoft.</td><td valign="top" rowspan="1" xmlns="">ETI Eloquence<a id="" name="">ETI Eloquence</a></td><td valign="top" rowspan="1" xmlns=""><br />rule-based formant-synthesis (Klatt-style). Later sold by Speechworks, also licensed to IBM (ViaVoice Outloud)</td><td valign="top" rowspan="1" xmlns="">DE, UK, US, ES, MX, FR, CA(FR), IT, FI, BR, CN, JP, KR</td><td valign="top" xmlns=""><br /></td><td valign="top" xmlns="">1998</td><td valign="top" xmlns=""><a href="ttsSamples/eloqdemo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/eloq_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td></tr><tr class="tablecell2"><td valign="top" rowspan="6"><a id="gdata" name="gdata" xmlns="">GData</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="6" xmlns="">Logox<a id="" name="">Logox</a></td><td valign="top" rowspan="6" xmlns="">microsegment synthesis<br />Microsegmentsynthesis (concatenating subphonetic units), not developed any more. Originally based on research from Univ. of Saarbrcken.</td><td valign="top" rowspan="6" xmlns="">DE, US, UK</td><td valign="top" xmlns="">Default voice<br /></td><td valign="top" xmlns="">2000</td><td valign="top" xmlns=""><a href="ttsSamples/logox_demo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td><tr class="tablecell2"><td valign="top" xmlns="">Bill<br /></td><td valign="top" xmlns="">1998</td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Bill (Swabian accent)<br /></td><td valign="top" xmlns="">2002</td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-schwab-s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-schwab-s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-schwab-s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Bill (Hessian accent)<br /></td><td valign="top" xmlns="">2002</td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-hessisch-s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-hessisch-s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-hessisch-s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Bill (Saxon accent)<br /></td><td valign="top" xmlns="">2002</td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-sachs-s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-sachs-s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-sachs-s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Bill (French accent)<br /></td><td valign="top" xmlns="">2002</td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-franz-s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-franz-s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/logox-bill-franz-s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="3"><a id="google" name="google" xmlns="">Google</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="2" xmlns="">Google Basic<a id="google-basic" name="google-basic" href="https://cloud.google.com/text-to-speech/">Google Basic</a></td><td valign="top" rowspan="2" xmlns="">so-called basic<br />The wavenet samples were not yet accessible for German.</td><td valign="top" rowspan="2" xmlns="">DE, NL, AUS, GB, US, CA, FR, JP, PT, SP, SE, TR</td><td valign="top" xmlns="">Standard A (female)<br /></td><td valign="top" xmlns="">2018</td><td valign="top" xmlns=""><a href="ttsSamples/google.basic-standard_A.s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/google.basic-standard_A.s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/google.basic-standard_A.s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Standard B (male)<br /></td><td valign="top" xmlns="">2018</td><td valign="top" xmlns=""><a href="ttsSamples/google.basic-standard_B.s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/google.basic-standard_B.s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/google.basic-standard_B.s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" rowspan="1" xmlns="">Google Translate<a id="google-translate" name="google-translate">Google Translate</a></td><td valign="top" rowspan="1" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="1" xmlns=""></td><td valign="top" xmlns="">Female<br />Samples were accessed via the translation service.</td><td valign="top" xmlns="">2013</td><td valign="top" xmlns=""><a href="ttsSamples/google_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/google_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/google_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2"><td valign="top" rowspan="2"><a id="ibm" name="ibm" href="https://www.ibm.com/watson/services/text-to-speech/" xmlns="">IBM</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="2" xmlns="">CTTS<a id="" name="">CTTS</a></td><td valign="top" rowspan="2" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="2" xmlns="">DE, US, UK, JP, KR, IT, ES, FR</td><td valign="top" xmlns="">Male<br />Courtesy of IBM. Database speaker is Gilles Karolyi. Sentence 3 sample is 8 kHz.</td><td valign="top" xmlns="">2002</td><td valign="top" xmlns=""><a href="ttsSamples/ctts_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/ctts_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/ctts_8kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" xmlns="">Female<br /><br /> Other sample: <a href="ttsSamples/ctts_female_8kHz.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">2004</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="4"><a id="infovox" name="infovox" xmlns="">Infovox</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="2" xmlns="">330/Infovox Desktop<a id="" name="">330/Infovox Desktop</a></td><td valign="top" rowspan="2" xmlns="">diphone-concatenation<br />Probably same as Babeltech Babil. Infovox 310 is apple version</td><td valign="top" rowspan="2" xmlns="">DE, UK, DK, NL, FI, FR, IS, IT, NO, ES, SE</td><td valign="top" xmlns="">Helga<br />8 kHz version<br /> Other sample: <a href="ttsSamples/telia330-1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">1996</td><td valign="top" xmlns=""><a href="ttsSamples/babil_helga_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/babil_helga_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/babil_helga_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Gerhard<br />8 kHz version<br /> Other sample: <a href="ttsSamples/infovox330_gerhard.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">1996</td><td valign="top" xmlns=""><a href="ttsSamples/babil_gerhard_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/babil_gerhard_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/babil_gerhard_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" rowspan="1" xmlns="">210/230<a id="" name="">210/230</a></td><td valign="top" rowspan="1" xmlns="">formant-synthesis<br />successor of KTH's OVE, originally telia promotor</td><td valign="top" rowspan="1" xmlns="">DE, UK, DK, NL, FI, FR, IS, IT, NO, ES, SE</td><td valign="top" xmlns=""><br /></td><td valign="top" xmlns="">1994</td><td valign="top" xmlns=""><a href="ttsSamples/telia230-1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/telia230-2_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td></tr><tr class="tablecell1"><td valign="top" rowspan="1" xmlns="">Desktop PRO<a id="" name="">Desktop PRO</a></td><td valign="top" rowspan="1" xmlns="">non-uniform unit-selection<br />same as Acapela HQ TTS</td><td valign="top" rowspan="1" xmlns=""></td><td valign="top" xmlns=""><br /></td><td valign="top" xmlns=""></td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td></tr></tr><tr class="tablecell2"><td valign="top" rowspan="4"><a id="innoetics" name="innoetics" href="http://www.innoetics.com/" xmlns="">Innoetics</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="4" xmlns=""><a id="" name=""></a></td><td valign="top" rowspan="4" xmlns="">non-uniform unit-selection<br />Development system from unsupervised audiobook extraction</td><td valign="top" rowspan="4" xmlns="">DE, US, UK, GR, BG</td><td valign="top" xmlns="">Christian<br />Courtesy of Innoetics</td><td valign="top" xmlns="">2015</td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Christian_01_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Christian_02_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Christian_03_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" xmlns="">Claudia<br />Courtesy of Innoetics</td><td valign="top" xmlns="">2015</td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Claudia_01_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Claudia_02_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Claudia_03_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Jessi<br />Courtesy of Innoetics</td><td valign="top" xmlns="">2015</td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Jessi_01_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Jessi_02_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Jessi_03_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Kalrsson<br />Courtesy of Innoetics</td><td valign="top" xmlns="">2015</td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Karlsson_01_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Karlsson_02_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/innoetics_Karlsson_03_b.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="2"><a id="ivona" name="ivona" href="http://www.ivona.com/" xmlns="">Ivona</a><br xmlns="" /><br xmlns="" />Owned by Amazon</td><td valign="top" rowspan="2" xmlns="">Ivona TTS<a id="" name="">Ivona TTS</a></td><td valign="top" rowspan="2" xmlns="">non-uniform unit-selection<br />Licensed by Lumenvox.</td><td valign="top" rowspan="2" xmlns="">DE, US, UK, ES, RO, PL, MX</td><td valign="top" xmlns="">Hans<br /></td><td valign="top" xmlns="">2011</td><td valign="top" xmlns=""><a href="ttsSamples/Ivona_Hans_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Ivona_Hans_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Ivona_Hans_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Marlene<br /></td><td valign="top" xmlns="">2011</td><td valign="top" xmlns=""><a href="ttsSamples/Ivona_Marlene_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Ivona_Marlene_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Ivona_Marlene_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2"><td valign="top" rowspan="2"><a id="lernouthauspie" name="lernouthauspie" xmlns="">Lernout &amp; Hauspie</a><br xmlns="" /><br xmlns="" />Acquired by Scansoft in 2001n after bankruptcy</td><td valign="top" rowspan="2" xmlns="">TTS3000<a id="" name="">TTS3000</a></td><td valign="top" rowspan="2" xmlns="">diphone</td><td valign="top" rowspan="2" xmlns="">DE, US, UK, NL, FR, RU, ES, MX, BR, CN, KR</td><td valign="top" xmlns="">Stefan<br /></td><td valign="top" xmlns="">1996</td><td valign="top" xmlns=""><a href="ttsSamples/lh_konkat.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/LHtts_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td><tr class="tablecell2"><td valign="top" xmlns="">Anna<br /><br /> Other sample: <a href="ttsSamples/lh_tts3000_anna.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">1996</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="3"><a id="loquendo" name="loquendo" xmlns="">Loquendo</a><br xmlns="" /><br xmlns="" />Acquired by Nuance in 2011</td><td valign="top" rowspan="3" xmlns="">Loquendo TTS<a id="" name="">Loquendo TTS</a></td><td valign="top" rowspan="3" xmlns="">non-uniform unit-selection<br />Formerly called Actor</td><td valign="top" rowspan="3" xmlns="">DE, IT, ES, FR, BR, PT, CN, UK, US, MX, GR, CL, AR, SE</td><td valign="top" xmlns="">Katrin<br />Courtesy of Loquendo.</td><td valign="top" xmlns="">2003</td><td valign="top" xmlns=""><a href="ttsSamples/loquendo_katrin_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/loquendo_katrin_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/loquendo_katrin_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Stefan<br />Courtesy of Loquendo.</td><td valign="top" xmlns="">2003</td><td valign="top" xmlns=""><a href="ttsSamples/loquendo_stefan_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/loquendo_stefan_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/loquendo_stefan_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Ulrike<br /></td><td valign="top" xmlns="">2001</td><td valign="top" xmlns=""><a href="ttsSamples/loquendo_ulrike_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/loquendo_ulrike_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/loquendo_ulrike_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2"><td valign="top" rowspan="1"><a id="meridian" name="meridian" href="http://speech.meridian-one.co.uk/orpheus.html" xmlns="">Meridian</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="1" xmlns="">Orpheus<a id="orpheus" name="orpheus">Orpheus</a></td><td valign="top" rowspan="1" xmlns="">formant<br />Formerly from Dolphin Oceanic Ltd. Specialized on fast speech as used by blind customers.</td><td valign="top" rowspan="1" xmlns="">DE, UK, US, FR, BR, PT, IT, ES, Welsh, CN, MD, CR, DN, NL, FI, GR, HU, LT, MY, NO, PL, RO, MX, SE</td><td valign="top" xmlns="">Orpheus<br /></td><td valign="top" xmlns="">2009</td><td valign="top" xmlns=""><a href="ttsSamples/orpheus_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/orpheus_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/orpheus_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" rowspan="1"><a id="microsoft" name="microsoft" xmlns="">Microsoft</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="1" xmlns="">Microsoft Speech Platform - Runtime Languages (Version 11)<a id="" name="">Microsoft Speech Platform - Runtime Languages (Version 11)</a></td><td valign="top" rowspan="1" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="1" xmlns="">ES, DK, DE, AU, CA, GB, IN, US, MX, FI, CA, FR, IT, JP, KR, NO, NL, PL, BR, PT, RU, SE, HK, TW, CN</td><td valign="top" xmlns="">Hedda<br /></td><td valign="top" xmlns="">2012</td><td valign="top" xmlns=""><a href="ttsSamples/ms_hedda_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/ms_hedda_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/ms_hedda_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" rowspan="2"><a id="neospeech" name="neospeech" href="http://neospeech.com/" xmlns="">Neospeech</a><br xmlns="" /><br xmlns="" />A Hoya company. As is ReadSpeaker.</td><td valign="top" rowspan="2" xmlns=""><a id="" name=""></a></td><td valign="top" rowspan="2" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="2" xmlns="">DE, US, UK, MX, TW, TH, KR, IT, CN, CH, JP, CT, BR, PT, FR</td><td valign="top" xmlns="">Lena<br /></td><td valign="top" xmlns="">2018</td><td valign="top" xmlns=""><a href="ttsSamples/Neospeech_lena_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Neospeech_lena_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Neospeech_lena_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" xmlns="">Tim<br /></td><td valign="top" xmlns="">2018</td><td valign="top" xmlns=""><a href="ttsSamples/Neospeech_tim_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Neospeech_tim_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Neospeech_tim_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="9"><a id="nuance" name="nuance" href="https://www.nuance.com" xmlns="">Nuance</a><br xmlns="" /><br xmlns="" />Formerly Scansoft (originating from Kurzweil and Xerox), acquired Europeean pioneers Lernout &amp; Hauspie in 2001, took the name of a smaller company named Nuance which they acquired in 2005 </td><td valign="top" rowspan="1" xmlns="">Vocalizer DNN<a id="" name="">Vocalizer DNN</a></td><td valign="top" rowspan="1" xmlns="">Artificial neural nets</td><td valign="top" rowspan="1" xmlns="">US</td><td valign="top" xmlns="">Nuance Website Sample<br /><br /> Other sample: <a href="ttsSamples/nuance-zoe-news-1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">2018</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td><tr class="tablecell1"><td valign="top" rowspan="8" xmlns="">Vocalizer<a id="" name="">Vocalizer</a></td><td valign="top" rowspan="8" xmlns="">non-uniform unit-selection<br />Formerly called RealSpeak (Vocalizer was the name of the original Nuance product), originally from Lernout &amp; Hauspie), converged with RVoice (formerly Rhetorical) . First commercial German unit-selection TTS</td><td valign="top" rowspan="8" xmlns="">DE, NL, PT, CA, CN, ES, DK, PT, FR, IT, JP, KR, MX, NO, PL, RU, SE, US, UK, AU, SA, ID, Basque, BE, CZ, FI, GR, IN, HU, TH, TR, ZA, RO</td><td valign="top" xmlns="">Victor<br /></td><td valign="top" xmlns="">2016</td><td valign="top" xmlns=""><a href="ttsSamples/vocalizer_victor_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/vocalizer_victor_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/vocalizer_victor_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Anna<br />11 kHz, courtesy of Nuance</td><td valign="top" xmlns="">2010</td><td valign="top" xmlns=""><a href="ttsSamples/realspeak_anna_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/realspeak_anna_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/realspeak_anna_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Yannick<br />11 kHz, courtesy of Nuance</td><td valign="top" xmlns="">2006</td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_yannik1.mp3&quot;"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_yannik2.mp3&quot;"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_yannik3.mp3&quot;"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Yannick 2<br />Yannick embedded version recorded from a cell phone</td><td valign="top" xmlns="">2009</td><td valign="top" xmlns=""><a href="ttsSamples/nuanceVocalizer_Yannick_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/nuanceVocalizer_Yannick_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/nuanceVocalizer_Yannick_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Monika<br />and Beate (?) - same as RVoice F026</td><td valign="top" xmlns="">2005</td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_f026_22kHz_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_f026_22kHz_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_f026_22kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Steffi<br />8 kHz</td><td valign="top" xmlns="">2004</td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_steffi1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_steffi2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_steffi3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Steffi 2<br />Newer version with enhanced voicequality and better pronunciation.</td><td valign="top" xmlns="">2005</td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_steffi_2-1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_steffi_2-2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_steffi_2-3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" xmlns="">Vera<br />8 kHz</td><td valign="top" xmlns="">1999</td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_vera1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_vera2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/realSpeak_vera3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr></tr><tr class="tablecell2"><td valign="top" rowspan="2"><a id="nuanceold" name="nuanceold" xmlns="">Nuance (until 2005)</a><br xmlns="" /><br xmlns="" />Acquired by Scansoft in 2005</td><td valign="top" rowspan="1" xmlns="">Vocalizer 4.05<a id="" name="">Vocalizer 4.05</a></td><td valign="top" rowspan="1" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="1" xmlns="">DE, US, UK, AU, CA(FR), MX*, BR</td><td valign="top" xmlns="">Anna Weber<br /></td><td valign="top" xmlns="">2004</td><td valign="top" xmlns=""><a href="ttsSamples/nuanceVocalizer_AnnaWeber_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/nuanceVocalizer_AnnaWeber_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td><tr class="tablecell2"><td valign="top" rowspan="1" xmlns="">Vocalizer 1.0<a id="" name="">Vocalizer 1.0</a></td><td valign="top" rowspan="1" xmlns="">non-uniform unit-selection<br />licensed Fonix engine</td><td valign="top" rowspan="1" xmlns="">DE, US, UK, NL, FR, IT, NO, ES, SE</td><td valign="top" xmlns=""><br /></td><td valign="top" xmlns="">2001</td><td valign="top" xmlns=""><a href="ttsSamples/nuance_1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/nuance_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="1"><a id="readspeaker" name="readspeaker" href="https://www.readspeaker.com/" xmlns="">ReadSpeaker</a><br xmlns="" /><br xmlns="" />A Hoya company. As is NeoSpeech. Formerly called rSpeak</td><td valign="top" rowspan="1" xmlns=""><a id="" name=""></a></td><td valign="top" rowspan="1" xmlns="">non-uniform unit-selection<br />using deep neural artificial networks</td><td valign="top" rowspan="1" xmlns="">DE,GB,US,AU,ES,FR,NL,SE</td><td valign="top" xmlns="">Max<br />courtesy of ReadSpeaker</td><td valign="top" xmlns="">2018</td><td valign="top" xmlns=""><a href="ttsSamples/ReadSpeaker_max_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/ReadSpeaker_max_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/ReadSpeaker_max_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" rowspan="3"><a id="rhetorical" name="rhetorical" xmlns="">Rhetorical Systems</a><br xmlns="" /><br xmlns="" />Was headquartered in Edinburgh, Scotland. Acquired by Scansoft / Nuance in 2004</td><td valign="top" rowspan="3" xmlns="">RVoice<a id="" name="">RVoice</a></td><td valign="top" rowspan="3" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="3" xmlns="">DE, UK, US, GR, ES</td><td valign="top" xmlns="">F026<br /></td><td valign="top" xmlns="">2004</td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_f026_22kHz_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_f026_22kHz_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_f026_22kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" xmlns="">M027<br /></td><td valign="top" xmlns="">2004</td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_m027_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_m027_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_m027_16kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">F018<br /></td><td valign="top" xmlns=""></td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_f018_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_f018_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/rhetorical_f018_16kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="1"><a id="speechworks" name="speechworks" xmlns="">Speechworks</a><br xmlns="" /><br xmlns="" />Acquired by Scansoft / Nuance in 2003</td><td valign="top" rowspan="1" xmlns="">Speechify<a id="" name="">Speechify</a></td><td valign="top" rowspan="1" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="1" xmlns="">DE, US, UK, AU, JP, MX*, FR, BR, CA(FR)</td><td valign="top" xmlns="">Tessa<br /></td><td valign="top" xmlns="">2002</td><td valign="top" xmlns=""><a href="ttsSamples/speechify_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/speechify_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/speechify_22kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" rowspan="4"><a id="svox" name="svox" xmlns="">Svox</a><br xmlns="" /><br xmlns="" />Originally a spin-off from ETH Zurich. Acquired by Nuance in 2011</td><td valign="top" rowspan="3" xmlns="">Svox Corporate<a id="" name="">Svox Corporate</a></td><td valign="top" rowspan="3" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="3" xmlns="">DE, FR, IT, US, ES</td><td valign="top" xmlns="">Petra<br /></td><td valign="top" xmlns="">2005</td><td valign="top" xmlns=""><a href="ttsSamples/svox_petra_22kHz_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/svox_petra_22kHz_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/svox_petra_22kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" xmlns="">Markus<br /></td><td valign="top" xmlns="">2005</td><td valign="top" xmlns=""><a href="ttsSamples/svox_markus_22kHz_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/svox_markus_22kHz_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/svox_markus_22kHz_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" xmlns="">Marlene<br /><br /> Other sample: <a href="ttsSamples/svoxUnitSel.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">2003</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td><td valign="top" xmlns="">-</td></tr><tr class="tablecell2"><td valign="top" rowspan="1" xmlns=""><a id="" name=""></a></td><td valign="top" rowspan="1" xmlns="">diphone</td><td valign="top" rowspan="1" xmlns="">DE, FR, IT, US, ES</td><td valign="top" xmlns="">Nicole<br /></td><td valign="top" xmlns="">2000</td><td valign="top" xmlns=""><a href="ttsSamples/svoxCommercial.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/svoxCommercial_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns="">-</td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="2"><a id="onscreenvoices" name="onscreenvoices" href="http://www.onscreenvoices.com/" xmlns="">Tom Weber Software</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="2" xmlns="">OnScreenVoices<a id="" name="">OnScreenVoices</a></td><td valign="top" rowspan="2" xmlns="">non-uniform unit-selection</td><td valign="top" rowspan="2" xmlns="">DE</td><td valign="top" xmlns="">Andreas<br />Samples courtesy of tom weber software</td><td valign="top" xmlns="">2015</td><td valign="top" xmlns=""><a href="ttsSamples/onscreenvoices_andreas_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/onscreenvoices_andreas_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/onscreenvoices_andreas_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" xmlns="">Marianne<br />Samples courtesy of tom weber software</td><td valign="top" xmlns="">2015</td><td valign="top" xmlns=""><a href="ttsSamples/onscreenvoices_marianne_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/onscreenvoices_marianne_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/onscreenvoices_marianne_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2"><td valign="top" rowspan="2"><a id="voiceinterconnect" name="voiceinterconnect" href="http://voiceinterconnect.de" xmlns="">VoiceINTERConnect</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="2" xmlns=""><a id="" name=""></a></td><td valign="top" rowspan="2" xmlns="">diphone<br />Commercial version of the Dress Synthesizer (University of Dresden).</td><td valign="top" rowspan="2" xmlns=""></td><td valign="top" xmlns="">female voice<br /></td><td valign="top" xmlns="">2000</td><td valign="top" xmlns=""><a href="ttsSamples/voiceInterConnectDress_w_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/voiceInterConnectDress_w_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/voiceInterConnectDress_w_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" xmlns="">male voice<br /></td><td valign="top" xmlns="">2000</td><td valign="top" xmlns=""><a href="ttsSamples/voiceInterConnectDress_m_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/voiceInterConnectDress_m_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/voiceInterConnectDress_m_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell1"><td valign="top" rowspan="1"><a id="votrax" name="votrax" href="http://en.wikipedia.org/wiki/Votrax" xmlns="">Votrax</a><br xmlns="" /><br xmlns="" /></td><td valign="top" rowspan="1" xmlns=""><a id="" name=""></a></td><td valign="top" rowspan="1" xmlns="">formant<br />Early hardware Formant synthesizer. Samples taken from an Audiodata Braille reader.</td><td valign="top" rowspan="1" xmlns="">DE</td><td valign="top" xmlns=""><br /></td><td valign="top" xmlns="">1974</td><td valign="top" xmlns=""><a href="ttsSamples/votrax_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/votrax_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/votrax_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top" rowspan="2"><a id="voxygen" name="voxygen" href="http://voxygen.fr" xmlns="">Voxygen</a><br xmlns="" /><br xmlns="" />Spin-off from French Orange Labs.</td><td valign="top" rowspan="2" xmlns=""><a id="" name=""></a></td><td valign="top" rowspan="2" xmlns="">Hybrid non-uniform unit-selection / HMM synthesis</td><td valign="top" rowspan="2" xmlns="">DE, FR, EN, ES, IT, AR</td><td valign="top" xmlns="">Sylvia<br />courtesy of Voxygen</td><td valign="top" xmlns="">2014</td><td valign="top" xmlns=""><a href="ttsSamples/Voxygen_s1_Sylvia.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Voxygen_s2_Sylvia.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Voxygen_s3_Sylvia.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" xmlns="">Matthias<br />courtesy of Voxygen</td><td valign="top" xmlns="">2014</td><td valign="top" xmlns=""><a href="ttsSamples/Voxygen_s1_Sean.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Voxygen_s2_Sean.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top" xmlns=""><a href="ttsSamples/Voxygen_s3_Sean.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr></table><h3><b>Universities / Research</b><a name="universities" id="universities" /></h3><table border="0"><tr class="tableheader"><th width="10%"><b>Institution</b></th><th width="10%"><b>System</b></th><th width="25%"><b>Remark</b></th><th width="25%">Year (approx.) / remark</th><th width="10%">s1</th><th width="10%">s2</th><th width="10%">s3</th></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="2"><a id="" name="">IKP Bonn</a><br /><br /></td><td valign="top" rowspan="1">BOSS</td><td valign="top" rowspan="1"><br />non-uniform unit-selection</td><td valign="top">2001<br /><br /></td><td valign="top"><a href="ttsSamples/boss_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/boss_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/boss_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" rowspan="1">Hadifix</td><td valign="top" rowspan="1"><br />mixed inventory concatenation<br />HADIFIX = HAlbsilben, DIphone und suFIXe<br />DE</td><td valign="top">1995<br /><br /></td><td valign="top"><a href="ttsSamples/hadifix_demo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/hadifix_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td></tr></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="3"><a id="" name="">University of Budapest</a><br /><br /></td><td valign="top" rowspan="2">Multivox 5 (Profivox)</td><td valign="top" rowspan="2"><br />diphone synthesis</td><td valign="top">2004<br />male speaker 1<br /></td><td valign="top"><a href="ttsSamples/profivox_spr1_s1.mp"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/profivox_spr1_s2.mp"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><tr class="tablecell2"><td valign="top">2004<br />male speaker 2<br /></td><td valign="top"><a href="ttsSamples/profivox_spr2_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/profivox_spr2_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td></tr><tr class="tablecell2"><td valign="top" rowspan="1">Multivox 3</td><td valign="top" rowspan="1"><br />formant synthesis<br />DE, HU, FI, NL, ES, PT, SA, Esperanto</td><td valign="top">1994<br /><br /><br /> Other sample: <a href="ttsSamples/multivox.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><td valign="top">-</td><td valign="top">-</td></tr></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="6"><a id="" name="" href="http://www.dfki.de/">DFKI</a><br /><br /></td><td valign="top" rowspan="5">Mary</td><td valign="top" rowspan="5"><br />non-uniform unit-selection<br />Mary=modular architecture for speech synthesis, open source. Great tool also to teach about speech synthesis because the output and input of different poicessing modules can be viewed as text.<br />DE, EN , Tibetian</td><td valign="top">2011<br />Pavoque corpus<br /></td><td valign="top"><a href="ttsSamples/pavoque_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/pavoque_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/pavoque_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top">2007<br />Bits 1<br />for details see Schrder, M. &amp; Hunecke, A. (2007). Creating German Unit Selection Voices for the MARY TTS Platform from the BITS Corpora. Proc. SSW6, Bonn, Germany.</td><td valign="top"><a href="ttsSamples/mary-bits1-s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/mary-bits1-s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/mary-bits1-s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top">2007<br />Bits 2<br /></td><td valign="top"><a href="ttsSamples/mary-bits2-s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/mary-bits2-s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/mary-bits2-s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top">2007<br />Bits 3<br /></td><td valign="top"><a href="ttsSamples/mary-bits3-s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/mary-bits3-s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/mary-bits3-s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top">2007<br />Bits 4<br /></td><td valign="top"><a href="ttsSamples/mary-bits4-s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/mary-bits4-s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/mary-bits4-s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top" rowspan="1">Mary/Mbrola</td><td valign="top" rowspan="1"><br />diphone<br />DE, EN</td><td valign="top">2000<br /><br /></td><td valign="top"><a href="ttsSamples/mary_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/mary_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/mary_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="6"><a id="" name="">Technical university of Dresden</a><br /><br /></td><td valign="top" rowspan="1">DRESS</td><td valign="top" rowspan="1"><br />diphone synthesis</td><td valign="top">1996<br /><br /></td><td valign="top"><a href="ttsSamples/dress1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/dress2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/dress3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top" rowspan="1">Voice 1</td><td valign="top" rowspan="1"><br />concatenative formant-synthesizer</td><td valign="top">1993<br /><br /><br /> Other sample: <a href="ttsSamples/v1demo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><td valign="top">-</td><td valign="top">-</td></tr><tr class="tablecell2"><td valign="top" rowspan="1">TUSY</td><td valign="top" rowspan="1"><br />hardware formant-synthesizer</td><td valign="top">1987<br /><br /><br /> Other sample: <a href="ttsSamples/tusy.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><td valign="top">-</td><td valign="top">-</td></tr><tr class="tablecell2"><td valign="top" rowspan="1">ROSY</td><td valign="top" rowspan="1"><br />hardware formant-synthesizer<br />Robotron Synthesizer</td><td valign="top">1977<br /><br /><br /> Other sample: <a href="ttsSamples/rosy.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><td valign="top">-</td><td valign="top">-</td></tr><tr class="tablecell2"><td valign="top" rowspan="1">Syni 2</td><td valign="top" rowspan="1"><br />punchcard controlled formant-synthesizer<br />Robotron Synthesizer</td><td valign="top">1975<br /><br /><br /> Other sample: <a href="ttsSamples/syni.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><td valign="top">-</td><td valign="top">-</td></tr><tr class="tablecell2"><td valign="top" rowspan="1">Syni 1</td><td valign="top" rowspan="1"><br />punchcard controlled formant-synthesizer<br />Robotron Synthesizer</td><td valign="top">1972<br /><br /><br /> Other sample: <a href="ttsSamples/syni1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><td valign="top">-</td><td valign="top">-</td></tr></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="" name="" href="http://espeak.sourceforge.net/">Jonathan Duddington</a><br /><br /></td><td valign="top" rowspan="1">eSpeak</td><td valign="top" rowspan="1"><br />formant-synthesis<br />based on the 1995 unix "speak"-program. Open-source</td><td valign="top">2006<br /><br /></td><td valign="top"><a href="ttsSamples/espeak-s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/espeak-s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/espeak-s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="" name="" href="http://www.tik.ee.ethz.ch/">ETH Zrich</a><br /><br /></td><td valign="top" rowspan="1">Svox</td><td valign="top" rowspan="1"><br />diphone-concatenation<br />Predecessor of the commercial version later acquired by Nuance.</td><td valign="top">1998<br /><br /></td><td valign="top"><a href="ttsSamples/svox_demo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/svox_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/svox_3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="" name="">Gerhard Mercator University of Duisburg</a><br /><br /></td><td valign="top" rowspan="1"></td><td valign="top" rowspan="1"><br />formant-synthesis</td><td valign="top">1996<br /><br /></td><td valign="top"><a href="ttsSamples/duisburg.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/duisburg_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="2"><a id="" name="" href="http://www.speech.kth.se/">KTH Stockholm</a><br /><br /></td><td valign="top" rowspan="1">Infovox</td><td valign="top" rowspan="1"><br />formant synthesis<br />Developed by Rolf Carlson, Bjorn Granstr;m and Sheri Hunnicut</td><td valign="top">1992<br /><br /></td><td valign="top"><a href="ttsSamples/kth_demo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><td valign="top">-</td><tr class="tablecell1"><td valign="top" rowspan="1">Ove III</td><td valign="top" rowspan="1"><br />Hardware formant synthesis<br />Orator Verbis Electris (OVE) . Developed by Gunnar Fant</td><td valign="top">1967<br /><br /><br /> Other sample: <a href="ttsSamples/oveIII.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><td valign="top">-</td><td valign="top">-</td></tr></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="8"><a id="" name="" href="http://tcts.fpms.ac.be/">University of Mons</a><br /><br /></td><td valign="top" rowspan="8">Mbrola</td><td valign="top" rowspan="8"><br />diphone-synthesis<br />Mbrola: Multi-band Resynthesis Overlap and Add. The NLP (text phonemisation) component is Txt2Pho, the Hadifix NLP in combination with Mbrola-Synthesis . Available for free for noncommercial use. MBROLA-TTS is avalable for about 34 different languages.</td><td valign="top">1998<br />de8<br />Markus Binsteiner's work an a Bavarian dialect<br /> Other sample: <a href="ttsSamples/binsteiner.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><td valign="top">-</td><td valign="top">-</td><tr class="tablecell2"><td valign="top">2000<br />de7<br />(by Marc Schrder, DFKI/Uni Saarland, female, 22 kHz), all diphones in three voice qualities (for emotional speech simulation).</td><td valign="top"><a href="ttsSamples/de7_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de7_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de7_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top">2000<br />de6<br />(by Marc Schrder, DFKI/Uni Saarland, male, 22 kHz), all diphones in three voice qualities (for emotional speech simulation).</td><td valign="top"><a href="ttsSamples/de6_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de6_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de6_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top">2000<br />de5<br />by Fred Englert (ATIP), female, 22 kHz</td><td valign="top"><a href="ttsSamples/de5_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de5_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de5_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top">2000<br />de4<br />By IMS Stuttgart, male, 16 kHz, includes english and french diphones</td><td valign="top"><a href="ttsSamples/de4_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de4_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de4_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top">2000<br />de3<br />by ATIP, female, first 22005 kHz voice</td><td valign="top"><a href="ttsSamples/de3_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de3_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de3_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top">1997<br />de2<br />By ATIP, male, 16 kHz</td><td valign="top"><a href="ttsSamples/de2_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de2_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de2_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top">1996<br />de1<br />By ATIP, female, 16 kHz</td><td valign="top"><a href="ttsSamples/de1_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de1_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/de1_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="" name="" href="http://www.ofai.at/">FAI (Austrian Research Institute for Artificial Intelligence)</a><br /><br /></td><td valign="top" rowspan="1">VieCtoS</td><td valign="top" rowspan="1"><br />demisyllable-LPC-concatenation<br />Vienna Concept-to-Speech system. If the prosody sounds poor it's due to my limited knowledge of Tobi-Labels.</td><td valign="top">1998<br /><br /></td><td valign="top"><a href="ttsSamples/viectos.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td><td valign="top">-</td></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="" name="">OGI, Oregon Graduate Institute, </a><br /><br /></td><td valign="top" rowspan="1"></td><td valign="top" rowspan="1"><br />LPC-diphone concatenation<br />Developed at the OGI, Center for Spoken Language Understanding during a workshop in 1998. TTS-Framework is Festival</td><td valign="top">1998<br /><br /></td><td valign="top"><a href="ttsSamples/ogiGerman.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/ogi_2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top">-</td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="" name="" href="http://www.ruhr-uni-bochum.de/">Ruhr Univeritt Bochum</a><br /><br /></td><td valign="top" rowspan="1">SyRUB, Version 4.1.1</td><td valign="top" rowspan="1" /><td valign="top">1995<br /><br /></td><td valign="top"><a href="ttsSamples/syrub_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/syrub_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/syrub_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="simple4all" name="simple4all" href="http://tundra.simple4all.org/demo/index.html">Simple4All</a><br /><br /></td><td valign="top" rowspan="1">Tundra corpus</td><td valign="top" rowspan="1"><br />non-uniform unit-selection<br />EU FP7 Project "Simple4All" Tundra corpus, system features unsupervised learning.</td><td valign="top">2013<br /><br /></td><td valign="top"><a href="ttsSamples/simple4all_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/simple4all_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><td valign="top"><a href="ttsSamples/simple4all_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></table><p><a name="others" id="others" /> with the following systems it wasn't possible to synthesize own sentences:</p><table border="0"><tr class="tableheader"><th width="10%"><b>name/link</b></th><th width="70%"><b>description</b></th><th width="10%">year (approx.)</th><th width="10%">mpeg3</th></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="2"><a id="aeg" name="aeg" href="http://de.wikipedia.org/wiki/Telefunken">AEG Telefunken</a><br /><br /></td><td valign="top" rowspan="1"><a id="spraus" name="spraus">SVS (SPRAUS Voll Synthese)</a><br /><br />unknown</td><td valign="top">1975<br /><br /></td><td valign="top"><a href="ttsSamples/aeg-spraus-svs.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top" rowspan="1"><a id="karlchen" name="karlchen">Karlchen</a><br /><br />unknown concatenation ("Parcor-Synthetisator")<br />Deutsche Bahn Auskunftssystem</td><td valign="top">1978<br /><br /></td><td valign="top"><a href="ttsSamples/karlchen.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="2"><a id="" name="" href="http://fastnet.netsoc.ie/chatr/">ATR</a><br /><br /></td><td valign="top" rowspan="2"><a id="" name=""></a><br /><br />non-uniform unit selection</td><td valign="top">1997<br />male<br /></td><td valign="top"><a href="ttsSamples/chatr_male.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell2"><td valign="top">1997<br />female<br /></td><td valign="top"><a href="ttsSamples/chatr_female.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="bose" name="bose">Bose</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name="">unkown</a><br /><br />unkown<br />recorded from a bose mini soundlink II bluetooth speaker february 2018</td><td valign="top">2018<br /><br /></td><td valign="top"><a href="ttsSamples/bose-speech-mini-feb2018.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="" name="">Univ. of Dresden, Peter Birkholz</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name="">Vocal Tract Lab</a><br /><br />Articulatory synthesis</td><td valign="top"><br /><br />Handtweaked articulatory movements transformed into a mathematical model to generate soundwaves</td><td valign="top">-</td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="2"><a id="" name="">ELIS Lab</a><br /><br /></td><td valign="top" rowspan="2"><a id="" name="">Eurovocs</a><br /><br />diphone-synthesis<br />Technology from Lernout &amp; Hauspie</td><td valign="top">1998<br /><br /></td><td valign="top"><a href="ttsSamples/eurovocs_demo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td><tr class="tablecell1"><td valign="top">1996<br /><br /></td><td valign="top"><a href="ttsSamples/eurovocsOld.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="firstbyte" name="firstbyte">First Byte</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name=""></a><br /><br />product-name:Monologue, ProVoice. waveform-concatenation synthesis (?</td><td valign="top">1998<br /><br /></td><td valign="top"><a href="ttsSamples/firstbyte_demo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="" name="" href="http://de.wikipedia.org/wiki/Fraunhofer-Institut_f%C3%BCr_Nachrichtentechnik#Geschichte">HHI: Heinrich Hertz Institut </a><br /><br /></td><td valign="top" rowspan="1"><a id="" name=""></a><br /><br />technology unknown</td><td valign="top">1978<br /><br /></td><td valign="top"><a href="ttsSamples/hhi.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="" name="">Keller &amp; Trauth.</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name="">SpeakEaZy</a><br /><br />waveform-concatenation synthesis</td><td valign="top">1998<br /><br /></td><td valign="top"><a href="ttsSamples/speakeazy.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="" name="">Wolfgang_von_Kempelen's Speaking Machine</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name=""></a><br /><br />Hardware manual sound generator ("papa", "mama")</td><td valign="top">1769<br /><br /></td><td valign="top"><a href="ttsSamples/kempelen.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="" name="">University of Kln</a><br /><br />Institut fr Phonetik</td><td valign="top" rowspan="1"><a id="" name=""></a><br /><br />articulatory-synthesis (actually not a TTS-system)</td><td valign="top">1996<br /><br /></td><td valign="top"><a href="ttsSamples/koeln_artikulator_demo.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="" name="" href="http://de.wikipedia.org/wiki/Karl_K%C3%BCpfm%C3%BCller">Karl Kpfmller / Bernhard Cramer</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name=""></a><br /><br />Hardware phoneme concatenation</td><td valign="top">1955<br /><br /></td><td valign="top"><a href="ttsSamples/kuepfmueller.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="" name="">University of Lausanne (LAIP)</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name=""></a><br /><br />TTS-system from the university of Lausanne (LAIP), uses MBROLA -engine. Includes a model to reduce/elaborate articulation according to speech-rate.</td><td valign="top">1998<br /><br /></td><td valign="top"><a href="ttsSamples/LAIP.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="mila" name="mila">Mila (Machine learning laboratory at the University of Montrea)</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name="">Char2Wav</a><br /><br />Deep neural artificial networks from University of Montreal: An end-to-end model for speech synthesis learned with Deeplearning4J. Char2Wav has two components: a reader and a neural vocoder. The reader is an encoder-decoder model with attention. The encoder is a bidirectional recurrent neural network that accepts text or phonemes as inputs, while the decoder is a recurrent neural network (RNN) with attention that produces vocoder acoustic features. For the German samples, the Pavoque database was used for training.</td><td valign="top">2017<br /><br /></td><td valign="top"><a href="ttsSamples/Char2Wav_best_bidirectional.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="" name="">Philips/IPO Eindhoven</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name="">Spengi</a><br /><br />diphone-synthesis</td><td valign="top">1997<br /><br /></td><td valign="top"><a href="ttsSamples/spengi.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="" name="">Unknown Russian TTS</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name=""></a><br /><br />unknown / formant?</td><td valign="top">1970<br /><br /></td><td valign="top"><a href="ttsSamples/russian.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="" name="">H.W. Strube, University of Gttingen</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name=""></a><br /><br />Articulatory synthesis.</td><td valign="top">1977<br /><br /></td><td valign="top"><a href="ttsSamples/strube.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="" name="">Texas Instruments Language Translator</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name=""></a><br /><br />LPC coded word-concatenation</td><td valign="top">1980<br />Male Voice<br /></td><td valign="top"><a href="ttsSamples/TexasInstrumentsLanguageTranslator_german.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2" xmlns=""><td valign="top" rowspan="1"><a id="" name="">University of West Bohemia in Pilsen</a><br /><br /></td><td valign="top" rowspan="1"><a id="" name="">ARTIC (ARtificial Talker In Czech)</a><br /><br />concatenative synthesizer<br />Commercial version available by speechtech by the name of ERIS.</td><td valign="top">2002<br /><br /></td><td valign="top"><a href="ttsSamples/artic.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1" xmlns=""><td valign="top" rowspan="1"><a id="" name=""></a><br /><br /></td><td valign="top" rowspan="1"><a id="" name=""></a><br /></td><td valign="top"><br /><br /></td><td valign="top">-</td></tr></table><hr /><a name="service" id="service" /><h3>Service products</h3><p> The following table lists some products to enhance text-to-speech quality.</p><table><tr class="tableheader"><th>company</th><th>product</th><th>description</th><th>date</th><th>sample</th></tr><tr class="tablecell1"><td valign="top"><a name="readSpeaker" id="readSpeaker" href="http://www.readspeaker.com">ReadSpeaker</a>, now commercialize their own engine under the name rSpeak, both a Hoya company.</td><td valign="top">SagEs / SayIt</td><td valign="top">Serverbased website reader. Based on <a href="#acapela">Acapela</a> products. Sample reads a newspaper article (Tagesspiegel). Note pronunciation of the word "playstation".</td><td valign="top">7/11/07</td><td valign="top"><a href="ttsSamples/readspeaker.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top"><a name="etex" id="etex" href="http://www.etex.de/">ETeX</a></td><td valign="top">-</td><td valign="top">Dictionaries.</td><td valign="top">1/7/05</td><td valign="top"><a href="ttsSamples/etex_dagmarGreeting.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top">Interlinx, aquired by <a href="http://www.speechconcept.com/">Speech Concept</a></td><td valign="top">emphasis / SpeechOptimizer</td><td valign="top">Tuning tool for pronounciation and prosody modeling.</td><td valign="top">1/7/05</td><td valign="top"><a href="ttsSamples/emphasis_weather.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></table><hr /><a name="other" id="other" /><h3>Further examples</h3><p> Speechsynthesis examples, that did not fit otherwise.</p><table><tr class="tableheader"><th>Description</th><th>Example</th></tr><tr class="tablecell1" name="ultraFast" id="ultraFast"><td valign="top">Ultrafast speechsynthesis as used by blind, with 14 syllables per second, based on formant synthesis <a href="#eloquent">Eloquence</a></td><td valign="top"><a href="ttsSamples/ultraFast_14ss.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top"><a href="#realspeak">realspeak</a><a href="http://www.scansoft.com/speechworks/realspeak/demo/default.asp">British English</a>, 31/5/05, "Flight LH312 from Frankfurt to Berlin."</td><td valign="top"><a href="ttsSamples/realspeak_fun1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell1"><td valign="top">TTS of the Fiat "Blue &amp; Me" Navigation Headunit with Microsoft CE. Voice Steffi of <a href="#nuance">Nuance</a>.</td><td valign="top"><a href="ttsSamples/microsoft_blueAndMe.wav"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr><tr class="tablecell2"><td valign="top">Apple Iphone 2011, Recorded with PC Mikrofon from Apple iPhone 4.1, TTS is faster compact version of Voice Yannick von <a href="#nuance">Nuance</a></td><td valign="top"><a href="ttsSamples/apple_s1.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a>, <a href="ttsSamples/apple_s2.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a>, <a href="ttsSamples/apple_s3.mp3"><img src="images/play.gif" alt="mp3" border="0" /></a></td></tr></table><hr /><a name="licensed" id="licensed" /><h3>Licensed Systems</h3><p> the following engines are based on systems with a different name:</p><ul><li><a href="http://www.mediainterface.de/">SpeaKing Synthesis</a> uses <a href="#svox">SVOX</a></li><li><a href="http://www.linguatec.de/products/tts/voice_reader">LinguaTec VoiceReader</a> uses <a href="#svox">SVOX Petra voice</a></li><li><a href="http://www.lumenvox.com/products/tts/#">Lumenvox</a> uses <a href="#ivona">Ivona</a></li><li><a href="http://www.tik.ee.ethz.ch/~spr/SPGinfo/node21.html">POSSY</a> from the ETH Zrich is a multilingual extension of <a href="#uni_svox">SVOX</a></li><li>Infovox Desktop Version 2.0 PRO same as Babeltech's Brightspeech</li><li><a href="http://www.windi7.com/voicepro.htm">VoicePro</a> from WinDi see Babeltech/Mbrola.</li><li>IBM's Viavoice Outloud see Eloquent.</li><li><a href="http://www.digalo.com/">Digalo</a> see Acapela Elan Tempo</li><li><a href="http://www.voicerss.org/">Voice RSS</a> uses Microsoft Hedda</li></ul><a name="missing" id="missing" /><h3>Missing examples</h3><p>For the following systems I didn't yet get samples:</p><ul><li><b>ALLVOC</b>, predecessor of Elan from France Telecom based on PSOLA</li><li><b>Papageno</b>, from Siemens</li><li><b>Tubsy</b> from Technical University Berlin</li><li><b>SyRUB</b> from the University of Bochum.<br /></li></ul><hr /><a name="unknown" id="unknown" /><h3>Unknown examples</h3><p>For the following systems I have no information about the supplier:</p><ul /><hr /><p><h3><a name="ttsChart" id="ttsChart">Categorization of text-to-speech systems</a></h3> Systems are usually either system- or signal modeling, primarily rule-based or data-based and can be distinguished by the type of the basic units and the way they are coded. <img src="images/TTSoverview.gif" border="0" alt="tts technology overview" /></p><hr /><a name="credits" id="credits" /><h3>Credits:</h3><p>The following persons delivered information and/or samples:</p><ul><li>Ulf Beckmann</li><li>Patrick Chabane</li><li>Bernhard Frtschl</li><li>Robert Kachel</li><li>Adrian Kurz</li><li>Michael Lang</li><li>Ali Savas</li><li>Stefan Seide</li><li>Bernhard Zeller</li></ul><hr /><a name="changelog" id="changelog" /><h3>Changelog</h3><ul><li>2018/4/24: added <a href="#google-basic">Google's basic samples</a> sample</li><li>2018/4/24: added <a href="#karlchen">AEG Karlchen sample</a> sample</li><li>2018/4/24: added <a href="#spraus">AEG SPRAUS sample</a> sample</li><li>2018/2/16: added <a href="#nuance">Nuance DNN</a> sample</li><li>2018/2/15: complete re-write in XML/XSLT, added <a href="#bose">Bose</a> sample</li><li>2018/1/3: added <a href="#readspeaker">new ReadSpeaker</a> samples and <a href="#neospeech">NeoSpeech</a>, both Hoya companies. Removed German version of this page. Too much work</li><li>2017/12/13: added <a href="#acapela">Author's custom voice</a></li><li>2017/05/12: added <a href="#char2wav">Char2Wav</a></li><li>2017/03/1: added <a href="#acapela">acapela</a> Claudia Smile sample</li><li>2017/02/15: added <a href="#readspeaker">rSpeak</a> Max samples</li><li>2016/10/28: added <a href="#nuance">Nuance vocalizer Victor</a> example phrase samples</li><li>2016/10/13: added <a href="#nuance">Nuance vocalizer Victor</a> sample</li><li>2016/05/09: linked second <a href="#firstbyte">Firstbyte Provoice</a> sample</li><li>2016/04/20: updated <a href="#aristech">Aristech</a> Alex samples</li><li>2016/1/4: added <a href="#votrax">Votrax</a> full samples</li><li>2015/12/4: removed VoiceRSS because they used Microsoft TTS</li><li>2015/12/4: added <a href="#nuance">Vocalizer (Realspek) Steffi newer version</a></li><li>2015/9/17: added <a href="#innoetics">Innoetics</a></li><li>2015/5/18: added <a href="#acapela">Acapela voice Claudia</a></li><li>2015/5/9: added <a href="#onscreenvoices">OnScreenVoices</a> samples</li><li>2014/12/18: added <a href="#strube">H.W. Strube</a> sample</li><li>2014/12/17: added <a href="#karlchen">Karlchen</a></li><li>2014/12/17: added <a href="#votrax">Votrax</a>, <a href="#kempelen">Von Kempelen</a>, <a href="#kuepfmueller">Kpfmller</a>, <a href="#hhi">HHI</a>, <a href="#aeg">AEG Telefunken</a>, <a href="#oveIII">OVE III</a> and <a href="#russian">unknown Russian</a> TTS</li><li>2014/11/27: added <a href="#voxygen">Voxygen</a></li><li>2014/9/25: removed Lumenvox</li><li>2014/1/29: added <a href="#acapela">Acapela child voices Lea and Jonas</a></li><li>2014/1/29: removed broken links</li><li>2013/10/29: added <a href="#google">Google</a></li><li>2013/09/18: added <a href="#simple4all">Simple4all</a></li><li>2013/09/5: added <a href="#lumenvox">Lumenvox</a></li><li>2013/08/15: changed SpeechConcept to <a href="#aristech">Aristech</a> and updated samples</li><li>2013/01/28: added <a href="#voicerss">VoiceRSS samples</a></li><li>2013/01/28: introduced <a href="#ivona">Amazon</a></li><li>2012/11/28: added <a href="#microsoft">Microsoft samples</a></li><li>2012/01/05: shifted identified samples to <a href="#other">other</a> section</li><li>2012/01/05: removed Apple because Nuance TTS is used</li><li>2012/01/05: added <a href="#acapela">Acapela voice Andreas samples</a></li><li>2012/01/03: added <a href="#syrub">SyRUB samples</a></li><li>2012/01/03: added <a href="#aristech">SpeechConcept Leopold voice samples</a></li><li>2011/09/08: added <a href="#unknown">unknown section.</a></li><li>2011/09/05: added <a href="#aristech">SpeechConcept corporate voice samples</a>. Noted aquisition of Svox and Loquendo by Nuance. Removed product comparison table (too much work to keep up-to-date)</li><li>2011/02/28: added <a href="#acapela">Acapela HQTTS Andreas</a> sample.</li><li>2011/02/22: added <a href="#mary">Mary Pavoque</a> samples.</li><li>2011/02/21: added <a href="#">Ivona</a> samples.</li><li>2011/02/16: added <a href="#apple">iPhone</a> samples.</li><li>2010/10/8: added <a href="#vocalizer">Vocalizer Anna</a> samples.</li><li>2010/01/06: added <a href="#nuance">Vocalizer Yannick embedded samples</a></li><li>2009/09/17: added <a href="#acapela">BrightSpeech Julia samples</a></li><li>2009/09/17: added <a href="#speechConcept">SpeechConcept samples</a></li><li>2009/09/17: added <a href="#meridian">orpheus from meridian</a></li><li>2009/01/15: added <a href="ttsProducts.xml">product comparison table</a></li><li>2008/09/23: added <a href="#berkom">berkom speech sample</a></li><li>2008/09/15: added <a href="#ultraFast">ultra fast speech sample</a></li><li>2008/04/14: added <a href="#acapela">Acapela greeting bunny sample</a></li><li>2007/11/07: added <a href="#readSpeaker">ReadSpeaker sample</a></li><li>2007/10/15: added <a href="#espeak">e-speak</a> and <a href="#vocalTractLab">vocal tract lab</a></li><li>2007/09/28: added <a href="#mary">Mary BITS samples</a></li><li>2007/07/14: added <a href="#texas">Texas Instruments Language Translator</a></li><li>2006/11/29: added new voice (Klaus) for <a href="#brightspeech">BrightSpeech</a></li><li>2006/8/21: added updated <a href="#loquendo">Loquendo samples</a></li><li>2006/6/19: added <a href="#vocalizer">Vocalizer Yannick</a> samples.</li><li>2005/11/3: added <a href="#vocalizer">vocalizer</a> sample.</li><li>2005/10/11: added second <a href="#laip">LAIP-TTS</a> sample.</li><li>2005/08/29: added <a href="#samt">SAMT</a> sample.</li><li>2005/07/01: added <a href="#etex">ETeX</a> and <a href="#interlinx">Interlinx</a> service descriptions.</li><li>2005/06/23: added <a href="#svox">SVOX</a> unit-selection samples, courtesy of SVOX.</li><li>2005/03/31: added realspeak <a href="#fun">fun sample</a></li><li>2005/02/22: added Logox accent samples</li><li>2005/01/04: added MBROLA de-X samples</li><li>2004/12/21: added Scansoft Steffi samples</li><li>October 29th 2004: added Chatr female sample</li><li>October 25th 2004: added Binsteiner sample</li><li>September 23rd 2004: added univ. stuttgart unit-selection sample</li><li>August 11th 2004: added more s3 samples</li><li>August 10th 2004: added Boss s2 sample</li><li>August 9th 2004: added rvoice F026 samples</li><li>July 7th 2004: added IBM ctts female sample</li><li>July 4th 2004: added artic synthesizer</li><li>June 24th 2004: added sentence 3 for selected commercial engines</li><li>June 22th 2004: added multivox 5 and old dresden formant-synthesizers</li><li>June 11th 2004: added voiceINTERConnect</li><li>June 10th 2004: added languages for commercial engines</li><li>June 10th 2004: added erkan and fifi voices for atip's proser</li></ul><hr /><p><a href="http://emosamples.syntheticspeech.de/index.html">Speechsynthesis-demos with simulated emotion</a></p><script type="text/javascript"> var pkBaseURL = (("https:" == document.location.protocol) ? "https://syntheticspeech.de/piwik/" : "http://syntheticspeech.de/piwik/"); document.write(unescape("%3Cscript src='" + pkBaseURL + "piwik.js' type='text/javascript'%3E%3C/script%3E")); </script><script type="text/javascript"> try { var piwikTracker = Piwik.getTracker(pkBaseURL + "piwik.php", 3); piwikTracker.trackPageView(); piwikTracker.enableLinkTracking(); } catch( err ) {} </script><noscript><p><img src="http://syntheticspeech.de/piwik/piwik.php?idsite=3" style="border:0" alt="" /></p></noscript></body></html>