can`t use word_timestamps it gives an error.

[furkanguzel161]

hello i am trying to use timestamps but i am getting this error.
its a modified version of whisper i modified the latest version from pip to make it compatible with intel gpu, its the same modification leuc has done: main...leuc:whisper:main
whisper test.mp3 --model large-v3 --language Turkish --output_format json --device xpu --word_timestamps True C:\Users\SlyRebula\miniconda3\envs\whisper\Lib\site-packages\torchvision\io\image.py:13: UserWarning: Failed to load image Python extension: 'Could not find module 'C:\Users\SlyRebula\miniconda3\envs\whisper\Lib\site-packages\torchvision\image.pyd' (or one of its dependencies). Try using the full path with constructor syntax.'If you don't plan on using image functionality from torchvision.io, you can ignore this warning. Otherwise, there might be something wrong with your environment. Did you have libjpegorlibpnginstalled before buildingtorchvision from source? warn( Traceback (most recent call last): File "C:\Users\SlyRebula\miniconda3\envs\whisper\Lib\site-packages\whisper\transcribe.py", line 481, in cli result = transcribe(model, audio_path, temperature=temperature, **args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\SlyRebula\miniconda3\envs\whisper\Lib\site-packages\whisper\transcribe.py", line 326, in transcribe add_word_timestamps( File "C:\Users\SlyRebula\miniconda3\envs\whisper\Lib\site-packages\whisper\timing.py", line 298, in add_word_timestamps alignment = find_alignment(model, tokenizer, text_tokens, mel, num_frames, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\SlyRebula\miniconda3\envs\whisper\Lib\site-packages\whisper\timing.py", line 205, in find_alignment weights = torch.stack([QKs[_l][_h] for _l, _h in model.alignment_heads.indices().T]) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ RuntimeError: indices expected sparse coordinate tensor layout but got Sparse Skipping test.mp3 due to RuntimeError: indices expected sparse coordinate tensor layout but got Sparse