search

clovaai / voxceleb_trainer

In defence of metric learning for speaker recognition
MIT License
1.03k stars 272 forks source link

Speed up Speed down #80

Closed hangxiu closed 3 years ago

hangxiu commented 3 years ago

self.speedup.build_array(input_array=audio, sample_rate_in=16000),0)

There is no specific implementation of speed interference in the paper. How can I change it if I use this enhancement?

joonson commented 3 years ago

This was from an unsuccessful experiment and should have been removed before. It has been removed now.