PL-EESR: Perceptual Loss Based END-TO-END Robust Speaker Representation Extraction
This addresses the issue of robust speaker representation extraction for speech processing applications, but it is incremental as it builds on existing methods with a hybrid approach.
The paper tackles the problem of speech distortion and speaker information loss in speech enhancement, which degrades speaker embedding extraction, by proposing an end-to-end framework called PL-EESR that uses perceptual loss and speaker identification feedback. The result shows better performance in speaker verification tasks in both clean and noisy environments compared to the baseline.
Speech enhancement aims to improve the perceptual quality of the speech signal by suppression of the background noise. However, excessive suppression may lead to speech distortion and speaker information loss, which degrades the performance of speaker embedding extraction. To alleviate this problem, we propose an end-to-end deep learning framework, dubbed PL-EESR, for robust speaker representation extraction. This framework is optimized based on the feedback of the speaker identification task and the high-level perceptual deviation between the raw speech signal and its noisy version. We conducted speaker verification tasks in both noisy and clean environment respectively to evaluate our system. Compared to the baseline, our method shows better performance in both clean and noisy environments, which means our method can not only enhance the speaker relative information but also avoid adding distortions.