Comparison of speech synthesizers

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

Here is a non-exhaustive comparison of speech synthesis programs:


Name Creator(s) First public release date Latest stable version Software license Cost
Alfanum TTS Alfanum 2004 2018 Proprietary ?
Apple PlainTalk Apple Inc. 1984 2018 Bundled with Mac OS X Bundled
AT&T Natural Voices AT&T Natural Voices ? 2008 Proprietary $295 – $995
Polly Amazon AWS 2016 2019 Proprietary $4.00 per 1 million characters (free in 1st year)
Cepstral Cepstral 2000 2013 Proprietary $29+
CereProc CereProc 2006 2017, February Proprietary £25.99+
CPqD Texto Fala CPqD 1999 2016, March 1st Proprietary ?
eSpeak Jonathan Duddington 2006, February 10 2010, December 2 GPLv3+ Free
Ekho Cameron Wong 2008, March 26 2021, April 06 GPLv2+ Free
Expressive Speech Voxygen SAS 2011, September ? Proprietary Not Free
Festival Speech Synthesis System CSTR ? 2014, December MIT-like license Free
FreeTTS Paul Lamere
Philip Kwok
Dirk Schnelle-Walka
Willie Walker
2001, December 14 2009, March 9 BSD Free
LumenVox LumenVox 2011 2019 Proprietary ?
Microsoft Speech API Microsoft 1995 2012 Bundled with Windows Bundled
VoiceText ReadSpeaker (Formerly Neospeech) 2002 2017 Proprietary ?
Nuance Vocalizer Nuance Communications, Inc. ? 2018 Proprietary Not Free
Praat Paul Boersma
David Weenink
1991 2021, February 28 GPL Free

Technical voice details[edit]

Platform SSML SAPI version WS PLS CLI
Alfanum TTS Yes 4.x/5.x ? ? ?
Apple PlainTalk ? ? ? ? ?
AT&T Natural Voices Yes 5.1 ? ? ?
Cepstral (company) Yes 5.x Yes Yes Yes
CereProc Yes 5.x Yes Yes Yes
CPqD Texto Fala Yes ? Yes ? Yes
Ekho ? ? ? ? ?
eSpeak Yes 5.x ? ? Yes
Expressive Speech 1.0/1.1 5.x ? Yes ?
Festival Speech Synthesis System ? ? ? ? Yes
FreeTTS ? ? ? ? ?
LumenVox Yes 5.x Yes Yes Yes
Microsoft Speech API 5.x only 4.x/5.x ? ? ?
Nuance Vocalizer ? ? ? ? ?
Praat ? ? ? ? ?
VoiceText Yes 5.x ? ? ?

Technical details[edit]

Name Online demo Available language(s) Available voices Programming language Operating system(s)
Alfanum TTS Yes Serbian, Croatian 8 C++ Windows
Apple PlainTalk ? English (United States), ... 15+ ? Macintosh
AT&T Natural Voices Yes English (British), English (Indian), English (US), French, French (Canadian), German, Italian, Spanish (Latin American) 20 C++ Linux
AWS Polly Yes Arabic (arb), Chinese, Mandarin (cmn-CN), Danish (da-DK), Dutch (nl-NL), English (Australian) (en-AU), English (British) (en-GB), English (Indian) (en-IN), English (US) (en-US), English (Welsh) (en-GB-WLS), French (fr-FR), French (Canadian) (fr-CA), German (de-DE), Hindi (hi-IN), Icelandic (is-IS), Italian (it-IT), Japanese (ja-JP), Korean (ko-KR), Norwegian (nb-NO), Polish (pl-PL), Portuguese (Brazilian) (pt-BR), Portuguese (European) (pt-PT), Romanian (ro-RO), Russian (ru-RU), Spanish (European) (es-ES), Spanish (Mexican) (es-MX), Spanish (US) (es-US), Swedish (sv-SE), Turkish (tr-TR), Welsh (cy-GB) 60
(male, female for most languages. For some languages child and different dialects are available)
Undisclosed by AWS Cloud-based online software with API adoptable for all currently available operating systems
Cepstral Yes English (British), English (US), Italian, French (Canadian), German, Spanish (American), ... 25+ C/C++ Mac OS X
CereProc Yes English (British), English (US), English (Scottish), English (Irish), French, French (Canadian), German, Austrian German, Italian, Irish, Spanish (Castilian), Spanish (Latin American), Dutch, Polish, Portuguese, Portuguese (Brazilian), Japanese, Catalan, Scottish Gaelic, Swedish, Russian, Mandarin 46

Java / C C++ / Objective C / Python / C# & .Net through SAPI

Mac OS X
Embedded Linux
Cloud service
CPqD Texto Fala Yes Brazilian Portuguese, Latin American Spanish, US English 5 C, C++ and Java Windows
Ekho Yes Cantonese, Mandarin (standard Chinese), Zhaoan Hakka (a dialect in Taiwan), Tibetan, Ngangien (an ancient Chinese before Yuan Dynasty) and Korean 7 C++ Linux
eSpeak Samples Afrikaans, Albanian, Armenian, Cantonese, Catalan, Croatian, Czech, Danish, Dutch, English (British, US, Scottish, Westindies...), Esperanto, Estonian, Finnish, French (France, Belgium), Georgian, German, Greek, Hindi, Hungarian, Icelandic, Indonesian, Italian, Kannada, Kurdish, Latvian, Lojban, Macedonian, Malayalam, Mandarin, Norwegian, Persian, Polish, Portuguese, Romanian, Russian, Serbian, Slovak, Spanish, Swahili, Swedish, Tamil, Turkish, Vietnamese, Welsh. Several C++ Linux
Mac OS X
Expressive Speech ? French, French (Canadian), French (African), UK English, US English, Spanish, German, Italian, Arabic, Wolof 50 C/C++/Java/Python Windows
Festival Speech Synthesis System Yes English (UK), English (US), Spanish, Hindi, Croatian, Finnish, Polish, Welsh. Several C++ Linux
FreeTTS ? English... Several Java Cross-platform
LumenVox Yes Danish, Dutch, English (Australian), English (US), English (UK), English (Welsh), English (Indian), French, French (Canadian), German, Icelandic, Italian, Polish, Portuguese, Portuguese (Brazilian), Romanian, Russian, Spanish (North American), Spanish (Latin American), Spanish (Castilian), Swedish, Turkish, Welsh, Welsh English 57 C/C++ Windows
Nuance Vocalizer Yes US English, Australian English, Indian English, Irish English, South African English, UK English, Argentinian Spanish, Castilian Spanish, Colombian Spanish, Mexican Spanish, Arabic, Catalan, Basque, Galician, Dutch, Belgian Dutch, Portuguese, Brazilian Portuguese, Bulgarian, French, Canadian French, Cantonese (Hong Kong), Mandarin, Mandarin Taiwanese, Czech, Danish, Finnish, German, Greek, Hebrew, Hindi, Hungarian, Indonesian, Italian, Japanese, Korean, Norwegian, Polish, Romanian, Russian, Slovak, Swedish, Thai, Turkish 70+ C/C++ Windows
Praat ? ? ? C Windows
VoiceText Yes English (US), English (British), American Spanish, Canadian French, Chinese Mandarin, Japanese, Korean 13 C/C++/Java Windows

Text-to-speech engines on Android[edit]

Name Creator(s) Available languages (voices) Latest version Last updated Google Play Store rating
Voxygen - Expressive Speech Voxygen 6 (50) 1.6.0 2017-12 varies (4-5)
Acapela TTS Voices Acapela Group 35 (100+) 2015-04-07 3.6 (2,111)
CereProc Text-to-Speech CereProc 11 (26) 4.0.5 2017-03 varies
Eloquence Text-To-Speech Nuance 10 1.2.0 2015-03-02 4.3 (129)
eSpeak eSpeak 40+ 1.46.02 2012-12-14 3.3 (1,762)
Google TTS Google 13 (16) ? 2015-04-07 4.0 (429,325)
IVONA Text-to-Speech HQ IVONA 13 (13) 2015-06-23 3.9 (14,413)
NeoSpeech NewsSpeak NeoSpeech 1 (7) 1.1.0 2014-12-30 4.8 (16)
SVOX Classic Text-to-Speech Nuance 25+ (40+) ? 2012-09-28 3.7 (15,740)
Vocalizer Nuance 36 (80+) 1.0.5 2015-04-23 3.5 (487)