search

taylorlu / Speaker-Diarization

speaker diarization by uis-rnn and speaker embedding by vgg-speaker-recognition
Apache License 2.0
469 stars 121 forks source link

How good is the ghostvlad weights of the model? #12

Closed fazlekarim closed 5 years ago

fazlekarim commented 5 years ago

How good are the weights to the ghostvlad model? Do I need to train it some more or the weights are the ones presented by the authors in their paper? Did you verify this?

taylorlu commented 5 years ago

Just a pretrained model, I haven't verify it, you had better retrain it.

fazlekarim commented 5 years ago

Did you use the weights from: https://github.com/WeidiXie/VGG-Speaker-Recognition or did you retrain it yourself on your re-implementation (https://github.com/taylorlu/ghostvlad-speaker)?

taylorlu commented 5 years ago

The 2 models' weights are the same, I haven't retrained.