cvqluu / simple_diarizer

Simplified diarization pipeline using some pretrained models - audio file to diarized segments in a few lines of code
GNU General Public License v3.0
141 stars 27 forks source link

Return 'cluster_labels' when 'extra_info=True' #11

Closed miguelvalente closed 1 year ago

miguelvalente commented 1 year ago

It would be handy if the labels for the embeddings were returned when requesting extra info.

cvqluu commented 1 year ago

Hi, this is a good idea. I think I'd prefer if extra_info is growing to instead return everything inside an dict.

So instead of additionally returning embeds, segments, cluster_labels separately, a dictionary containing each, i.e. {'embeds': embeds, 'segments': segments, ...}

Do you mind adding that and also updating the docstring?

miguelvalente commented 1 year ago

Hey, thanks, those labels would indeed be handy. Just made the extra changes as well. :v:

cvqluu commented 1 year ago

Looks good to me, merged. Will update the pip package

miguelvalente commented 1 year ago

Awesome. Thanks! :v: