Anyone Doing Speech To Text?
All;
I have a customer who is looking for a good speech to text solution, either open source or reasonably priced commercial product, I’m open to suggestions.
Thanks;
John V
All;
I have a customer who is looking for a good speech to text solution, either open source or reasonably priced commercial product, I’m open to suggestions.
Thanks;
John V
8 thoughts on - Anyone Doing Speech To Text?
hi you can try this link
http://zaf.github.io/asterisk-googletts/
2015-08-26 19:15 GMT+01:00 Tech Support:
For a commercial option try Lumenvox, had very good results. For “free”
you can try google tts but you never know when google will decide to pull the plug on something.
I had been using google tts, but it started requiring a captcha for my browser, and via linux I can’t access http://translate.google.com/translate_tts?q=test (redirects to captcha)
as so, its not reliable
I have a similar situation here, I want to include TTS in my asterisk IVR
system. Could someone give suggestion(s) please, I prefer open-source thanks in advance!
Chatila, A. C. P. O. Box 365, Kihesa Street, Njombe, Tanzania.
*Mob:* +255 765 154 235
*Whatsapp:* +255 653 258 608
*Website:* chax.me.tz
Hello,
what follows is a mostly incomplete list of Text To Speech (TTS) and Speech To Text (STT)
solutions available for asterisk.
-Regarding the TTS free and open source available options:
Asterisk comes with festival (http://www.cstr.ed.ac.uk/projects/festival/) support
(app_festival) already build in. Decent quality, supports mainly English.
There is support for flite (http://www.festvox.org/flite/) available as a 3rd party plugin : http://zaf.github.io/Asterisk-Flite/ Quality at par with festival, much easier to setup and use, supports only English.
Also support for espeak (http://espeak.sourceforge.net/):
http://zaf.github.io/Asterisk-eSpeak/ Average quality, supports a wide range of languages.
-Free plugins/scripts that provide TTS from a remote not-so-free service:
GoogleTTS : http://zaf.github.io/asterisk-googletts/ Great quality, lots of languages, free of charge but NOT suitable for any serious/commercial use. It is not a service Google officially provides but just a hack that gets synthesized speech data from their translate page. It’s more suitable for testing/developing and home use.
MsTTS: http://zaf.github.io/asterisk-mstts/ using Microsoft’s Translator voice synthesis engine.
iSpeech: http://zaf.github.io/asterisk-ispeech/ using iSpeech API (http://www.ispeech.org)
-Other non free solutions:
Cepstral: http://www.cepstral.com/en/telephony/asterisk
Speech Technology Group: http://www.asteriskexchange.com/listings/1001
-Regarding the STT options:
Google Speech: http://zaf.github.io/asterisk-speech-recog/ the API is limited at the moment in something like 50 requests/day and considered a technology preview.
iSpeech: http://zaf.github.io/asterisk-ispeech/
Lumevox: http://www.lumenvox.com/partners/digium/asterisk.aspx
Sphinx: http://cmusphinx.sourceforge.net/wiki/asteriskdetails
Vestec: http://www.asteriskexchange.com/listings/113
Regards,
Lefteris Zafiris
You’re right, I misinterpreted
Sorry!
Lefteris,
Thanks a lot for your detailed answer and for the valuable work you’ve been doing on this topic for quite some time now.
Cheers,
Philippe Sultan
2015-08-28 12:26 GMT+02:00 Lefteris Zafiris:
been doing on this topic for quite some time now.
+1
googletts-cli.pl is now my ‘go to’ for creating prompts for prototyping systems.