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 2108.12128cs.SDcs.AIeess.AS
收藏

基于任务的语音增强中的Warping因子

Task-aware Warping Factors in Mask-based Speech Enhancement

NEC Corporation(日本电气株式会社);Institute for Infocomm Research, A*STAR(新加坡科技研究局资讯通信研究院);Yokohama City University(横滨市立大学)

Qiongqiong Wang, Kong Aik Lee, Takafumi Koshinaka, Koji Okabe, Hitoshi Yamamoto

摘要:

本文提出了使用两个任务感知的翘曲因子在基于掩码的语音增强(SE)。一个控制在训练阶段中语音保持和噪声去除之间的平衡,而另一个控制在测试阶段中应用于特定下游任务的SE功率。我们的目的是减轻SE系统训练,以提高语音质量往往无法改善其他下游任务,如自动说话人验证(ASV)和自动语音识别(ASR),因为他们不共享相同的对象的问题。这是很容易应用所提出的双翘曲因子的方法,任何基于掩模的SE方法,它允许一个单一的SE系统来处理多个任务,而无需任务相关的培训。我们所提出的方法的有效性已被证实的SITW数据集的ASV评估和LibriSpeech数据集的ASR和语音质量评估的0- 20 dB。我们表明,单个SE需要不同的翘曲值才能实现最佳性能三项任务。使用任务相关的翘曲因子,语音质量提高了84.7%的PESQ增加,ASV有22.4%的EER减少,和ASR有52.2%的WER减少,在0 dB的语音。还在用于ASV的VoxCeleb-1测试集和用于ASV和质量评估的LibriSpeech dev-clean集上交叉验证了任务相关扭曲因子的有效性。所提出的方法是非常有效的,易于在实践中应用。

英文摘要:

This paper proposes the use of two task-aware warping factors in mask-based speech enhancement (SE). One controls the balance between speech-maintenance and noise-removal in training phases, while the other controls SE power applied to specific downstream tasks in testing phases. Our intention is to alleviate the problem that SE systems trained to improve speech quality often fail to improve other downstream tasks, such as automatic speaker verification (ASV) and automatic speech recognition (ASR), because they do not share the same objects. It is easy to apply the proposed dual-warping factors approach to any mask-based SE method, and it allows a single SE system to handle multiple tasks without task-dependent training. The effectiveness of our proposed approach has been confirmed on the SITW dataset for ASV evaluation and the LibriSpeech dataset for ASR and speech quality evaluations of 0-20dB. We show that different warping values are necessary for a single SE to achieve optimal performance w.r.t. the three tasks. With the use of task-dependent warping factors, speech quality was improved by an 84.7% PESQ increase, ASV had a 22.4% EER reduction, and ASR had a 52.2% WER reduction, on 0dB speech. The effectiveness of the task-dependent warping factors were also cross-validated on VoxCeleb-1 test set for ASV and LibriSpeech dev-clean set for ASV and quality evaluations. The proposed method is highly effective and easy to apply in practice.

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。

↑