Extract supervised model's speaker embeddings

sensein / senselab

SenseLab is a Python package that simplifies building pipelines for biometric (e.g. speech, voice, video, etc) analysis.

https://github.com/sensein/senselab

Apache License 2.0

6 stars 3 forks source link

Closed 900miles closed 2 weeks ago

900miles commented 3 weeks ago

Description: Implement functionality to extract speaker embeddings from a given tensor using any of xvector, ECAPA-TDNN, or ResNet.

From #25