search

philipperemy / deep-speaker

Deep Speaker: an End-to-End Neural Speaker Embedding System.
MIT License
905 stars 241 forks source link

How can I continue training with my own dataset #67

Closed VinhDevNguyen closed 4 years ago

VinhDevNguyen commented 4 years ago

Hi, Is it possible if I use your pretrained model to continue training with my own dataset? If it possible can you show me how to do that? Thanks !!!

philipperemy commented 4 years ago

@VinhDevNguyen yes def possible! All you need is to have the same format as the LibriSpeech dataset. Just create a big folder, and speakers as sub folders and put your wav/flac files in there. Then rebuild your caches: https://github.com/philipperemy/deep-speaker/blob/cc1a90baca155da824603f59bb3d54cef86f2152/deep-speaker#L29. And I think you can restart the commands and should work!

VinhDevNguyen commented 4 years ago

Thank you so much! I will try it ;)

philipperemy commented 4 years ago

Let me know ;)