# PL-EESR: Perceptual Loss Based END-TO-END Robust Speaker Representation Extraction

arXiv: https://arxiv.org/abs/2110.00940 · HTML: https://mlsanity.com/paper/2110.00940/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 27 references that are in the map.
- First posted on arXiv 3 Oct 2021.

## Builds on (most cited first, 10 of 10)

- [Perceptual Losses for Real-Time Style Transfer and Super-Resolution](https://mlsanity.com/paper/1603.08155/) (2016)
- [VoxCeleb2: Deep Speaker Recognition](https://mlsanity.com/paper/1806.05622/) (2018)
- [MUSAN: A Music, Speech, and Noise Corpus](https://mlsanity.com/paper/1510.08484/) (2015)
- [ICASSP 2021 Deep Noise Suppression Challenge](https://mlsanity.com/paper/2009.06122/) (2020)
- [VoiceID Loss: Speech Enhancement for Speaker Verification](https://mlsanity.com/paper/1904.03601/) (2019)
- [On Cross-Corpus Generalization of Deep Learning Based Speech Enhancement](https://mlsanity.com/paper/2002.04027/) (2020)
- [Speaker-Utterance Dual Attention for Speaker and Utterance Verification](https://mlsanity.com/paper/2008.08901/) (2020)
- [Feature Enhancement with Deep Feature Losses for Speaker Verification](https://mlsanity.com/paper/1910.11905/) (2019)
- [Robust Speaker Recognition Using Speech Enhancement And Attention Model](https://mlsanity.com/paper/2001.05031/) (2020)
- [Analysis of Deep Feature Loss based Enhancement for Speaker Verification](https://mlsanity.com/paper/2002.00139/) (2020)

## Built on by (most cited first, 2 of 2)

- [Disentangling Voice and Content with Self-Supervision for Speaker Recognition](https://mlsanity.com/paper/2310.01128/) (2023)
- [MFA: TDNN with Multi-scale Frequency-channel Attention for Text-independent Speaker Verification with Short Utterances](https://mlsanity.com/paper/2202.01624/) (2022)

## Abstract (arXiv)

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.