Encountering Value Error When Running the Provided Colab Example

akashmjn / tinydiarize

Minimal extension of OpenAI's Whisper adding speaker diarization with special tokens

MIT License

439 stars 14 forks source link

Encountering Value Error When Running the Provided Colab Example #18

Open prafulkl opened 1 year ago

prafulkl commented 1 year ago

Issue Description

I encountered an issue when using the code from the sample Colab file. The code is as follows:

model = whisper.load_model("small.en-tdrz")
result = whisper.transcribe(model, "demo_video.mp4", verbose=True, condition_on_previous_text=False)
# time segments are a bit more stable without prev_text conditioning as it wasn't yet fine-tuned that way

ValueError: tuple.index(x): x not in tuple

Also, I tried the model with MP3 and WAV files but got the same error!

akashmjn commented 1 year ago

I just tried the Colab notebook as well and run into this issue. Will look into what's up (likely has something to do with this PR https://github.com/akashmjn/tinydiarize/pull/11), thanks for pointing it out.