Hi,
For me the recognition is good but my problem is when I start the recognition system (implemented it for android) it will straight away start recognising without anything spoken. The enviroment I recorded it in and the enviroment I am testing is the same. Is there anyway I can fire the recognition system only when there is a voice of a speaker ?
I have my own language model + implementing it on android .
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
For me the recognition is good but my problem is when I start the recognition system (implemented it for android) it will straight away start recognising without anything spoken. The enviroment I recorded it in and the enviroment I am testing is the same. Is there anyway I can fire the recognition system only when there is a voice of a speaker ?
I have my own language model + implementing it on android .
Increase vad_threhsold with
config.setFloat("-vad_threshold", 3.0f);