I am implementing server side automatic speech recognition using Nuance server SDK. But none of the the sound is recognized by Nuance server.
It is giving error 'Query retry 1 or 2' or 'server error 500'.
I have checked my audio. Neither it is silent, not its extension is wrong. I have tried .amr, .wav, .mp3 etc versions of audio.
What can be done for the same?
Or any api in php or javascript or cordova which converts recorded speech audio to text and supports mainly indian languages (hindi, gujarati etc) ?
