Support for stereo audio

2025-07-04 13:37:07 +02:00 · 2025-07-04 13:37:07 +02:00 · 6966635499
commit 6966635499
parent f8110e3487
1 changed files with 1 additions and 0 deletions
--- a/scripts/stt_from_file_pytorch.py
+++ b/scripts/stt_from_file_pytorch.py
@ -152,6 +152,7 @@ def main(args):

    audio, input_sample_rate = sphn.read(args.in_file)
    audio = torch.from_numpy(audio).to(args.device)
+    audio = audio.mean(dim=0, keepdim=True) # Convert to mono
    audio = julius.resample_frac(audio, input_sample_rate, mimi.sample_rate)
    if audio.shape[-1] % mimi.frame_size != 0:
        to_pad = mimi.frame_size - audio.shape[-1] % mimi.frame_size