Hi @wegylexy ,
could you explain to me, why you scale the input up and output down before and after the rnnoise_process_frame()?
And thank you for your work.
Edit: I guess it's because AudioWorklet sends 128 Frames 32-bit via the process() function. For rnnoise 16-bit is required. So the scaling is the conversion? Could you explain it in detail for me.
Hi @wegylexy , could you explain to me, why you scale the input up and output down before and after the rnnoise_process_frame()?
And thank you for your work.
Edit: I guess it's because AudioWorklet sends 128 Frames 32-bit via the process() function. For rnnoise 16-bit is required. So the scaling is the conversion? Could you explain it in detail for me.