MRL No sound from speech.py script



MRL No sound from speech.py script. The files are created in the audioFile directory but are 3kb and do not play.

Any Ideas?

RZ

If I copy the URL from the JAVA log and paste it into IE, I get audio. It seems to me the audio files are corrupt. I have tried deleting the files and letting the script recreate them but get same result.

GroG's picture

Hello and Welcome

Hello and Welcome rz90208,

Yes,  Google decided not to support its "unofficial" interface - so Speech is currently borked.
Fortunately - we found a great Open Source - replacement .. Yay - 
It's called MaryTTS - we are working to get it integrated with MRL now :)

Until then we may be a little mute ..

rz90208's picture

Thank you for the update

Thank you for the update. Not sure what direction Google is heading, the dropping of the Google code and now this.

bensonofjohn's picture

@Grog have you seen this

@Grog have you seen this solution? http://stackoverflow.com/questions/32053442/google-translate-tts-api-blocked

GroG's picture

Thanks bensonofjohns ! You

Thanks bensonofjohns !

You saved the GoogleSpeech service - we are in the process of refactoring so all script just need to use the "Speech" service ...

So far - for the back end we have now 3 possiblities - (previously GoogleSpeech was headed to the scrap pile)

When you use Speech - the it will proxy one of these 3:

  • (Default) MarySpeech - great Open Source project with very active development, multi-language, multi-voice support.  Does not sound as "smooth" as some of the commercial services - but allows far more potential as you can use the Source Luke !
  • AcapelaSpeech - a commercial online backend which offers demo for non commercial applications - Better sounding and more voices than Google
  • GoogleSpeech - recently broke there API - thanks to bensonofjohn might still be in service.