Vikhrmodels
/

Borealis

Automatic Speech Recognition

text-generation

Model card Files Files and versions

Anonumous commited on Sep 11, 2025

Commit

ff3cbdb

·

verified ·

1 Parent(s): e0a1c22

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -49,7 +49,7 @@ proc = extractor(
     )
 mel = proc.input_features.squeeze(0).to(device)
 with torch.inference_mode():
     transcript = model.generate(mel=mel, att_mask=att_mask, **generation_params)

     )
 mel = proc.input_features.squeeze(0).to(device)
+att_mask = proc.attention_mask.squeeze(0).to("cuda")
 with torch.inference_mode():
     transcript = model.generate(mel=mel, att_mask=att_mask, **generation_params)