Open chamecall opened 4 years ago
If you record audio yourself you can simply feed it into the recognizer, see https://github.com/alphacep/kaldi-android-demo/blob/83b23cfc34dde4b688fbfff7566a460bc9b56455/app/src/main/java/org/kaldi/demo/KaldiActivity.java#L132 method
I guess he wants to do the opposite : extract recognized audio and process it for an other use case (like saving audio AND text together)
Are there any updates with this issue? Or maybe someone else added this function?
Are there any updates on this issue. Is it possible to get audio file from the default recognizer used by Vosk or no?
Are there any updates on this issue. Is it possible to get audio file from the default recognizer used by Vosk or no?
Yes, it's possible. You can read in raw PCM data from android's AudioRecord and output it to a valid WAV file. You'd have to implement in SpeechService class and build..
Are there any updates on this issue. Is it possible to get audio file from the default recognizer used by Vosk or no?
Yes, it's possible. You can read in raw PCM data from android's AudioRecord and output it to a valid WAV file. You'd have to implement in SpeechService class and build..
Can you send example?
Did you create your own implementation of SpeechService? Because inside original implementation AudioRecord field is private.
Did you create your own implementation of SpeechService? Because inside original implementation AudioRecord field is private.
Yes I did!
@nanaghartey could you share your fork please? I think the whole community will be grateful to you for this.
I had to refuse android.SpeechRecognizer because of the impossibility to extract audio from it so I'd like to find out how I can reach that purpose in here..