Closed ubanning closed 1 year ago
Thanks for bug report - gonna review ASAP.
Thanks for bug report - gonna review ASAP.
Ok, thanks a lot for the quick reply, I sent you an email too with basically the same message, sorry.
Thanks 😊
Obviously, you already changed the script to get rid of the bug. But now, I still get an error at step 6. It now says that the variable "result_merged" is not defined:
NameError Traceback (most recent call last)
[<ipython-input-17-01c09ad3d31b>](https://localhost:8080/#) in <module>
1 #@title Create SRT with transcription
2 result_srt_list = []
----> 3 for i, v in enumerate(result_merged):
4 result_srt_list.append(srt.Subtitle(index=i, start=timedelta(seconds=v['start']), end=timedelta(seconds=v['end']), content=v['text'].strip()))
5
NameError: name 'result_merged' is not defined
Another error I get is at step 2.2: "Select and Load Model"
ImportError Traceback (most recent call last)
<ipython-input-5-c9ab7d1f63d9> in <module>
1 #@title Step 2.2: Select and load Model.
2 if vad_filter:
----> 3 from whisperx.vad import VADSegmentPipeline
4 vad_pipeline = VADSegmentPipeline(model_name = vad_model,
5 device = device,
ImportError: cannot import name 'VADSegmentPipeline' from 'whisperx.vad' (/usr/local/lib/python3.9/dist-packages/whisperx/vad.py)
As a result, I had to turn off the vad filter.
Hello, first thanks for the work. 😊
I have an error when I execute step number 6, more precisely in the cell:
"Create SRT with transcription"
The error I get is:
Can you help me?
Another thing I noticed is that in step 3: Prepare audio for transcription
An example is provided for converting the audio:
ffmpeg -i aci.mp4 -vn -c:a aci.wav
but if you run just that command, a warning appears in the console:At least one output file must be specified
So I modified it to:
ffmpeg -i aci.mp4 -vn -c:a pcm_s16le aci.wav
or just:ffmpeg -i aci.mp4 aci.wav
I don't know if this modification has anything to do with the error I got.Thanks.