-
Hi, I have the same question as https://github.com/microsoft/SpeechT5/issues/16#issuecomment-1516257038. My training dataset is Chinese, so can i use speechbrain/spkrec-xvect-voxceleb to extract speak…
-
Hey you, thank you for the package :)
I'm researching around how to improve diarization errors related to overlapping speech, and I'd like to ask you about your choice of a embedding model.
Is t…
-
Since the typechecked specification changed in some version (I have not been able to find it in the documentation), it may be necessary to modify related to that in [many places](https://github.com/se…
-
Hi, thanks for sharing
I wonder how the pre-trained model was trained. What data did you use to train the encoder ? LibriSpeech ? VoxCeleb ?
-
### Describe the feature
I tried the different models for speaker ID listed here: https://github.com/k2-fsa/sherpa-onnx/releases/tag/speaker-recongition-models
And for example, using "[wespeaker_e…
-
Dear, author, thanks for your great work. when I try to train the model, it have many warning like this: error to parse id10536/aZT_gTu6ilw/00005.wav.wav. I check the voxceleb dataset seems ok, did yo…
-
When the first epoch ends, I've an error in the evaluation because there are some speakers/labels that are not present on label_encoder.txt
```
(.venv) root@8ad8297faf3e:/home/diarization/speechbr…
-
Is there dataloader for voxceleb?
-
thanks for ur working, i find the demo need voxceleb video .csv file. can u test using a custom vidoe without .csv file or using other landmark detections code to make a easy and generic demo?
-
Hi, the truth is that I am very new to the field of deep learning, so I have had a lot of trouble being able to retrain the model from its pre-trained model with the VoxCeleb dataset. I only have a Ge…