speaker detection

#1
by synthetisoft - opened

Any idea if there is a separate model which could automate the task of specifying the number of speakers prior to diarization? I know there are entirely separate diarization models which don't require you to specify but I am yet to have luck with them. Your space works well in contrast to the others.

Great input, thanks! Would definitely improve the app overall. I'll have a look into it whether there are any open-source models already available to include.

Sign up or log in to comment