Multilingual speaker embeddings

[sedol1339]

Hello! pyannote/speaker-diarization-3.1 uses wespeaker-voxceleb-resnet34-LM embedding model, however according to wespeaker docs it is only for English. Is there a multilingual embedding model for SpeakerDiarization constructor?

[hbredin]

It has been trained on mostly English data but will generalize quite well to other languages.

[sedol1339]

Thank you, will try it