arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2311.11599eess.AS
收藏

端到端语音识别训练如何影响语音增强伪影?

How does end-to-end speech recognition training impact speech enhancement artifacts?

Kazuma Iwamoto, Tsubasa Ochiai, Marc Delcroix, Rintaro Ikeshita, Hiroshi Sato, Shoko Araki, Shigeru Katagiri

AI总结:

研究联合训练语音增强前端与ASR后端对信号伪影的影响,发现ASR级训练减少伪影但增加噪声误差,且简单插值增强与观测信号即可改善ASR性能。

摘要:

联合训练语音增强(SE)前端和自动语音识别(ASR)后端已被研究作为减轻单通道SE产生的处理失真对ASR影响的一种方法。本文从分解的噪声和伪影误差的角度,研究这种联合训练对增强信号信号级特性的影响。实验分析提供了两个新发现:1)SE前端的ASR级训练减少了伪影误差,同时增加了噪声误差;2)简单地对增强信号和观测信号进行插值,即可达到类似减少伪影和增加噪声的效果,无需联合修改SE和ASR模块即可改善ASR性能,即使对于使用WavLM特征提取器的强大ASR后端也是如此。我们的发现为理解联合训练的效果提供了更好的认识,并为设计ASR不可知的SE前端提供了新见解。

英文摘要:

Jointly training a speech enhancement (SE) front-end and an automatic speech recognition (ASR) back-end has been investigated as a way to mitigate the influence of \emph{processing distortion} generated by single-channel SE on ASR. In this paper, we investigate the effect of such joint training on the signal-level characteristics of the enhanced signals from the viewpoint of the decomposed noise and artifact errors. The experimental analyses provide two novel findings: 1) ASR-level training of the SE front-end reduces the artifact errors while increasing the noise errors, and 2) simply interpolating the enhanced and observed signals, which achieves a similar effect of reducing artifacts and increasing noise, improves ASR performance without jointly modifying the SE and ASR modules, even for a strong ASR back-end using a WavLM feature extractor. Our findings provide a better understanding of the effect of joint training and a novel insight for designing an ASR agnostic SE front-end.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑