A Joint Training Framework of Multi-Look Separator and Speaker Embedding Extractor for Overlapped Speech
Naijun Zheng, Na Li, Bo Wu, Meng Yu, Jianwei Yu, Chao Weng, Dan Su, Xunying Liu
Abstract
In multi-talker cases, overlapped speech degrades the speaker verification (SV) performance dramatically. To tackle this challenging problem, speech separation with multi-channel techniques can be adopted to extract each speaker’s signals to improve the SV performance. In this paper, a joint training framework of the front-end multi-look speech separator and the back-end speaker embedding extractor is proposed for multi-channel overlapped speech. To better leverage the complementarity between the speech separator and the speaker embedding extractor, several training strategies are proposed to jointly optimize the two modules. Experimental results show that the proposed joint training framework significantly outperforms the individual SV system by around 52% relative EER reduction. Additionally, the robustness of the proposed framework is further evaluated under different conditions.
BibTeX
@inproceedings{icassp2021_ajointtrainingfr,
title = {A Joint Training Framework of Multi-Look Separator and Speaker Embedding Extractor for Overlapped Speech},
author = {Naijun Zheng and Na Li and Bo Wu and Meng Yu and Jianwei Yu and Chao Weng and Dan Su and Xunying Liu and Helen Meng},
booktitle = {ICASSP 2021},
year = {2021}
}