Datasets:
Transcripts misaligned / wrong
#7
by RonanMcGovern - opened
Quite a few transcripts appear wrong, based on test set inspection. The audio doesn't align to the text.
Can you please give me a list of utterances wrongly aligned? Thanks a lot!
Unfortunately I think the vast majority are quite misaligned, at least in the test set, I didn't look at train. For example, if you run whisper v3 large over the dataset you see a strong bimodal distribution.
