Puyuan Peng

flashattention might help

could you elaborate?

does it happen with the demo example? If not, it could be some error when you specify `target_transcript` and `edit_type` in cell 5 of `inference_speech_editing.ipynb`. for example, right now the...

Or you can reply here your `orig_transcirpt`, `target_transcript`, and `edit_type`.

Is it happening at ```python concat_sample = audio_tokenizer.decode( [(concat_frames, None)] # [1,T,8] -> [1,8,T] ) gen_sample = audio_tokenizer.decode( [(gen_frames, None)] ) ``` that means the model generated too few tokens....

> @jasonppy Hi, I should be ready on my side Thanks, I'll test it in the next two days

what is the error that you encounterred?

could you access this page? https://huggingface.co/pyp1

ah sorry about that, a while ago I updated the files to be safetensors, they should serve the same purpose right?

well, `.pth` got replaced with '' ``` model = voicecraft.VoiceCraft.from_pretrained(f"pyp1/VoiceCraft_{voicecraft_name.replace('.pth', '')}") ```