How to handle multiple speakers?

HarryVolek / PyTorch_Speaker_Verification

PyTorch implementation of "Generalized End-to-End Loss for Speaker Verification" by Wan, Li et al.

BSD 3-Clause "New" or "Revised" License

576 stars 164 forks source link

How to handle multiple speakers? #60

Open chrisspen opened 4 years ago

chrisspen commented 4 years ago

The config.yaml file is hardcoded to train the system for 4 speakers, even though the TIMIT data appears to contain many more speakers than this. Why is that? How would you use this to train an embedding system for an arbitrary number of speakers?