arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2309.02459cs.SDcs.CLeess.AS
收藏

通过下采样声学表征实现端到端语音识别的纯文本域自适应

Text-Only Domain Adaptation for End-to-End Speech Recognition through Down-Sampling Acoustic Representation

Tsinghua University(清华大学);Peng Cheng Lab(鹏城实验室);Tencent AI Lab(腾讯人工智能实验室);The Chinese University of Hong Kong(香港中文大学)

Jiaxu Zhu, Weinan Tong, Yaoxun Xu, Changhe Song, Zhiyong Wu, Zhao You, Dan Su, Dong Yu, Helen Meng

AI总结:

针对语音与文本表征长度不一致导致纯文本域自适应效果不佳的问题,提出下采样声学表征对齐文本模态的策略,引入CIF模块生成匹配词元长度的声学表征,提升端到端ASR的新域性能。

摘要:

将语音和文本这两种模态映射到共享表征空间,是利用纯文本数据提升新域下端到端自动语音识别(ASR)性能的研究方向。但语音表征与文本表征的长度不一致,现有方法通过上采样文本表征以对齐声学模态,可能与预期的实际时长不匹配。本文提出一种新的表征匹配策略:下采样声学表征以对齐文本模态。通过引入连续积分放电(CIF)模块生成与词元长度一致的声学表征,我们的ASR模型能更好地从两种模态中学习统一表征,从而可利用目标域的纯文本数据进行域自适应。新域数据的实验结果证明了所提方法的有效性。

英文摘要:

Mapping two modalities, speech and text, into a shared representation space, is a research topic of using text-only data to improve end-to-end automatic speech recognition (ASR) performance in new domains. However, the length of speech representation and text representation is inconsistent. Although the previous method up-samples the text representation to align with acoustic modality, it may not match the expected actual duration. In this paper, we proposed novel representations match strategy through down-sampling acoustic representation to align with text modality. By introducing a continuous integrate-and-fire (CIF) module generating acoustic representations consistent with token length, our ASR model can learn unified representations from both modalities better, allowing for domain adaptation using text-only data of the target domain. Experiment results of new domain data demonstrate the effectiveness of the proposed method.

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。

↑