Ask Ubuntu is a question and answer site for Ubuntu users and developers. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

If I'm developing an Android app that uses TTS and Voice recognition, which libraries are used for the same voice recognition and speech on Ubuntu?

I'm assuming espeak for text to speech, but I'm unsure which voice recognition library and dictionary/learning/calibration system is used for voice recognition.

I'ld like to make the app available on Ubuntu Desktop.
as well as test it outside an emulator

share|improve this question
check this but though there is no substantial answer – devav2 Sep 19 '12 at 11:43
@devav2 thanks for the effort, but not really related. voice login and which voice recognition toolkit is used on android (and if it is available in Ubuntu) are two completely different things. – RobotHumans Sep 19 '12 at 21:15
You'll probably need to find out exactly what libraries Android uses elsewhere, then come back here and ask how to use them in Ubuntu. If you can't find it in Android documentation, Android Enthusiasts or Stack Overflow might be of help. – ændrük Oct 2 '12 at 17:05
Not sure about using it with Android, but Festival/festvox sounds much less robotic than espeak. As for the speech recognition,… might help. – Adam Feb 15 '13 at 21:15
Well, espeak comes standard on Ubuntu desktop no dependency added. On the other hand, the paid packages from cepstral are not horrifyingly expensive and they border on skeksi. – RobotHumans Feb 16 '13 at 0:46
up vote 5 down vote accepted

You may use Google's webspeech API for your application. It uses the same technology as used on android.

share|improve this answer
If you edit in a little more data, I'll give this one the accept as well – RobotHumans Mar 24 '13 at 1:27
Honestly, I've not had enough experience with their API to give you that info! – ishitor Mar 27 '13 at 18:33
This might be an interesting complement: A bare-bones project to bring google's speech-to-text to my desktop. – Stéphane Gourichon Dec 17 '15 at 18:52
  1. Androids voice recognition relies on an internet connection, because the server has a substantial database of voice patterns. (I think)

    • Computers arent very good with voice recognition yet, so you need a huge database with alot of voices for the computer to cross reference and better understand words. (At least last i studied it)

    • As so, you cannot use it offline.

  2. You can however use the voice recognition for writing via a workaround

    • By creating a remote desktop/vnc connection, and direct

    • or by using a remote keyboard application.

Chrome can use voice recognition for searches, and with an extension, can be used to fill out field data. but as far as i know thats it.

and that's all i know.

share|improve this answer
I understand that voice recognition requires a corpus to learn words beyond simple phonetics and dictionaries (so fuzzy matches can be made when you don't enunciate). I'm fine with requiring an internet connection if it's usable. – RobotHumans Feb 23 '13 at 16:50

Maybe you use CMU Sphinx, it can be used as a toolkit to develop speech recognition both in Linux and Android: CMU Sphinx

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.