Can I get speaker embeddings via AM-MobileNet1D?

joaoantoniocn / AM-MobileNet1D

The Additive Margin MobileNet1D is a new light weight deep learning model for Speaker Recognition which is based on the MobileNetV2 architecture and the Additive Margin Softmax (AM-Softmax) loss function.)

29 stars 10 forks source link

Can I get speaker embeddings via AM-MobileNet1D? #7

Open mardaayden opened 3 years ago

mardaayden commented 3 years ago

Hi,

I am working on an offline speaker recognition task and I need to enroll people by vector embeddings without training the network again.

I know that the AM-MobileNet1D model is not for speaker embeddings however is there any way to get speaker embeddings via AM-MobileNet1D? For example; can I get last layer output before classification as vector embeddings?

Thanks

joaoantoniocn commented 3 years ago

Hi, you can try to access the features before classification, it may work as an embedding.

mardaayden commented 3 years ago

Thanks for your answer! I will try it and report the results here.