Closed Aurora11111 closed 5 years ago
These two libraries are for speaker embedding learning, not for diarization. They solve completely different problems than UIS-RNN. Please read the corresponding papers first.
For more details about these two repos, please contact their authors directly.
TensorFlow implementation by Janghyun1230 PyTorch implementaion by HarryVolek - with UIS-RNN integration hello,Are the two versions above have the same algorithm with uis-rnn ? what's the different between them ? Can all of them identify enrollment utterances and unenrollment utterances? Can all of them identify new speakers which is untrained ?