Speaker-Distinguishable CTC: Learning Speaker Distinction Using CTC for Multi-Talker Speech Recognition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sakuma, Asahi, Sato, Hiroaki, Sugano, Ryuga, Kumano, Tadashi, Kawai, Yoshihiko, Ogawa, Tetsuji
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908399467233280
author Sakuma, Asahi
Sato, Hiroaki
Sugano, Ryuga
Kumano, Tadashi
Kawai, Yoshihiko
Ogawa, Tetsuji
author_facet Sakuma, Asahi
Sato, Hiroaki
Sugano, Ryuga
Kumano, Tadashi
Kawai, Yoshihiko
Ogawa, Tetsuji
contents This paper presents a novel framework for multi-talker automatic speech recognition without the need for auxiliary information. Serialized Output Training (SOT), a widely used approach, suffers from recognition errors due to speaker assignment failures. Although incorporating auxiliary information, such as token-level timestamps, can improve recognition accuracy, extracting such information from natural conversational speech remains challenging. To address this limitation, we propose Speaker-Distinguishable CTC (SD-CTC), an extension of CTC that jointly assigns a token and its corresponding speaker label to each frame. We further integrate SD-CTC into the SOT framework, enabling the SOT model to learn speaker distinction using only overlapping speech and transcriptions. Experimental comparisons show that multi-task learning with SD-CTC and SOT reduces the error rate of the SOT model by 26% and achieves performance comparable to state-of-the-art methods relying on auxiliary information.
format Preprint
id arxiv_https___arxiv_org_abs_2506_07515
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Speaker-Distinguishable CTC: Learning Speaker Distinction Using CTC for Multi-Talker Speech Recognition
Sakuma, Asahi
Sato, Hiroaki
Sugano, Ryuga
Kumano, Tadashi
Kawai, Yoshihiko
Ogawa, Tetsuji
Audio and Speech Processing
Computation and Language
Sound
This paper presents a novel framework for multi-talker automatic speech recognition without the need for auxiliary information. Serialized Output Training (SOT), a widely used approach, suffers from recognition errors due to speaker assignment failures. Although incorporating auxiliary information, such as token-level timestamps, can improve recognition accuracy, extracting such information from natural conversational speech remains challenging. To address this limitation, we propose Speaker-Distinguishable CTC (SD-CTC), an extension of CTC that jointly assigns a token and its corresponding speaker label to each frame. We further integrate SD-CTC into the SOT framework, enabling the SOT model to learn speaker distinction using only overlapping speech and transcriptions. Experimental comparisons show that multi-task learning with SD-CTC and SOT reduces the error rate of the SOT model by 26% and achieves performance comparable to state-of-the-art methods relying on auxiliary information.
title Speaker-Distinguishable CTC: Learning Speaker Distinction Using CTC for Multi-Talker Speech Recognition
topic Audio and Speech Processing
Computation and Language
Sound
url https://arxiv.org/abs/2506.07515