Can I read X-SEPFORMER: End-To-End Speaker Extraction Network with Explicit Optimization on Speaker Confusion on EtoBox?

X-SEPFORMER: End-To-End Speaker Extraction Network with Explicit Optimization on Speaker Confusion by Kai Liu; Ziqing Du; Xucheng Wan; Huan Zhou is a scholarly article available to read on EtoBox.

What is X-SEPFORMER: End-To-End Speaker Extraction Network with Explicit Optimization on Speaker Confusion about?

Target speech extraction (TSE) systems are designed to extract target speech from a multi-talker mixture. The popular training objective for most prior TSE networks is to enhance reconstruction performance of extracted speech waveform. However, it has been reported that a TSE system delivers high reconstruction performance may still suffer lowquality experience problems in practice. One such experience problem is wrong speaker extraction (called speaker confusion, SC), which leads to strong negative experience and hampers effective conversations. To mitigate the imperative SC issue, we reformulate the training objective and propose two novel loss schemes that explore the metric of reconstruction improvement performance defined at small chunk-level and leverage the metric associated distribution information. Both loss schemes aim to encourage a TSE network to pay attention to those SC chunks based on the said distribution information. On this basis, we present X-SepFormer, an end-to-end TSE model with proposed loss schemes and a backbone of Sep-Former. Experimental results on the benchmark WSJ0-2mix dataset validate the effectiveness of our proposals, showing consistent improvements

Author

Kai Liu; Ziqing Du; Xucheng Wan; Huan Zhou

Publisher

IEEE

Published

2023

Language

EN