arXiv 2306.022732023-06-04cs.CLcs.SDeess.AS

端到端联合目标与非目标说话人自动语音识别

End-to-End Joint Target and Non-Target Speakers ASR

NTT Computer and Data Science Laboratories, NTT Corporation（NTT公司NTT计算机与数据科学实验室）

Ryo Masumura, Naoki Makishima, Taiga Yamane, Yoshihiko Yamazaki, Saki Mizuno, Mana Ihori, Mihiro Uchida, Keita Suzuki, Hiroshi Sato, Tomohiro Tanaka, Akihiko Ta… 展开作者

Ryo Masumura, Naoki Makishima, Taiga Yamane, Yoshihiko Yamazaki, Saki Mizuno, Mana Ihori, Mihiro Uchida, Keita Suzuki, Hiroshi Sato, Tomohiro Tanaka, Akihiko Takashima, Satoshi Suzuki, Takafumi Moriya, Nobukatsu Hojo, Atsushi Ando

备注：: Accepted at Interspeech 2023

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对多说话人重叠语音需同时转写目标与非目标说话人的问题，提出利用目标说话人注册语音并递归生成文本及说话人身份标记的端到端自回归ASR方法，实验验证了其有效性。

摘要：

本文提出一种新颖的自动语音识别（ASR）系统，能够从多说话人重叠语音中转写单个说话人的语音，同时识别其是目标说话人还是非目标说话人。目标说话人ASR系统通过注册目标说话人信息，仅转写目标说话人语音，是一种很有前景的方法。然而，在对话式ASR应用中，为理解交互信息，通常既需要转写目标说话人的语音，也需要转写非目标说话人的语音。为在单一ASR模型中自然地同时考虑目标与非目标说话人，本文的思路是扩展基于自回归建模的多说话人ASR系统，使其利用目标说话人的注册语音。所提出的ASR通过递归生成文本标记以及表示目标或非目标说话人的标记来执行。实验结果证明了所提方法的有效性。

英文摘要：

This paper proposes a novel automatic speech recognition (ASR) system that can transcribe individual speaker's speech while identifying whether they are target or non-target speakers from multi-talker overlapped speech. Target-speaker ASR systems are a promising way to only transcribe a target speaker's speech by enrolling the target speaker's information. However, in conversational ASR applications, transcribing both the target speaker's speech and non-target speakers' ones is often required to understand interactive information. To naturally consider both target and non-target speakers in a single ASR model, our idea is to extend autoregressive modeling-based multi-talker ASR systems to utilize the enrollment speech of the target speaker. Our proposed ASR is performed by recursively generating both textual tokens and tokens that represent target or non-target speakers. Our experiments demonstrate the effectiveness of our 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。