Transcription of mp3 repeats something about subtitles

[matthiasroder]

I have downloaded a video from youtube with youtube-dl

youtube-dl --extract-audio --audio-format mp3 -o "%(id)s.%(ext)s" https://www.youtube.com/watch\?v\=sAzxw4XVLUM

I then checked the resulting mp3 and it sounded fine.

After that I use whisper to transcribe, using this command:

whisper sAzxw4XVLUM.mp3 --language German

which gives me the following output:

/Users/matthias/opt/anaconda3/envs/ai/lib/python3.9/site-packages/whisper/transcribe.py:114: UserWarning: FP16 is not supported on CPU; using FP32 instead warnings.warn("FP16 is not supported on CPU; using FP32 instead") [00:00.000 --> 00:02.000] Untertitel im Auftrag des ZDF, 2017 [00:30.000 --> 00:40.000] Untertitel im Auftrag des ZDF [01:00.000 --> 01:20.000] Untertitel im Auftrag des ZDF [01:30.000 --> 01:40.000] Untertitel im Auftrag des ZDF [02:00.000 --> 02:28.000] Untertitel im Auftrag des ZDF [02:30.000 --> 02:50.000] Untertitel im Auftrag des ZDF ...

The m3 begins with silence, then music and then talking.

Any ideas what could cause the problem?

Many thanks!
Matthias

[mayeaux]

Check this thread out: #928

[matthiasroder]

Thank you, this was a very helpful hint.