format of sound data (mono/stereo, 2 bytes, 4 bytes)

[Ehsanwww]

Hello @cjheath, thanks for great asset, in order to detect text from voice or a sound clip, that data that should be passed into whisper_full() method, should be in what format? I mean mono/stereo and also 2 bytes or 4 bytes for each samples?

[Alumniminium]

16khz mono, 2 bytes (16bit) - also known as pcm_s16le

[cjheath]

Hello @cjheath, thanks for great asset

I am not an author here. Please don't ping other users.