search

GreatJiweix / DmmlTiSV

Deep multi-metric learning for text-independent speaker verification
24 stars 2 forks source link

problem of test_model for myself dataset. #1

Open juanting opened 4 years ago

juanting commented 4 years ago

Hello, thank you very much for sharing your excellent project. I have just started the field of voice print recognition, and I have seen your excellent open source code. I have a question want to ask you, I need a system to generate the embedding of speaker, found that you provide the pre-train model, I want to use this model to test my own dataset eer, I only have speak id and the corresponding voice clips. I want to ask next how to realize my validation? Do I need the file like veri_test.txt? Looking forward to your reply.

giusarno commented 3 years ago

Hello, I have a similar question but I do not seem to be able to download the weights. Is there a way I can download them ?