search

alphacep / vosk-android-demo

Offline speech recognition for Android with Vosk library.
Apache License 2.0
714 stars 187 forks source link

The audio stream with a frame rate of 44100 and dual channels is inaccurate #211

Open mabeijianxi opened 9 months ago

mabeijianxi commented 9 months ago

I tried to use a network audio stream with a frame rate of 44100 and dual channels for audio-to-text conversion, but the result was incorrect. I did not see any options for setting single or dual channels in the API. How should I solve this problem?

nshmyrev commented 9 months ago

You have to convert stereo to mono yourself, there is no code for that yet

mabeijianxi commented 8 months ago

Alright, thank you. May I ask if there is a plan to support the stereo format?

nshmyrev commented 8 months ago

Sorry, we do not have plans for this. Its not the task of the library, more an external tool. For example if you are using ffmpeg you can convert to mono.