Saved in:
Bibliographic Details
Main Authors: Yifan, Guo, Yao, Tian, Hongbin, Suo, Yulong, Wan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.02688
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911356385492992
author Yifan, Guo
Yao, Tian
Hongbin, Suo
Yulong, Wan
author_facet Yifan, Guo
Yao, Tian
Hongbin, Suo
Yulong, Wan
contents With the development of teleconferencing and in-vehicle voice assistants, far-field multi-speaker speech recognition has become a hot research topic. Recently, a multi-channel transformer (MCT) has been proposed, which demonstrates the ability of the transformer to model far-field acoustic environments. However, MCT cannot encode high-dimensional acoustic features for each speaker from mixed input audio because of the interference between speakers. Based on these, we propose the multi-channel multi-speaker transformer (M2Former) for far-field multi-speaker ASR in this paper. Experiments on the SMS-WSJ benchmark show that the M2Former outperforms the neural beamformer, MCT, dual-path RNN with transform-average-concatenate and multi-channel deep clustering based end-to-end systems by 9.2%, 14.3%, 24.9%, and 52.2% respectively, in terms of relative word error rate reduction.
format Preprint
id arxiv_https___arxiv_org_abs_2601_02688
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Multi-channel multi-speaker transformer for speech recognition
Yifan, Guo
Yao, Tian
Hongbin, Suo
Yulong, Wan
Sound
Artificial Intelligence
With the development of teleconferencing and in-vehicle voice assistants, far-field multi-speaker speech recognition has become a hot research topic. Recently, a multi-channel transformer (MCT) has been proposed, which demonstrates the ability of the transformer to model far-field acoustic environments. However, MCT cannot encode high-dimensional acoustic features for each speaker from mixed input audio because of the interference between speakers. Based on these, we propose the multi-channel multi-speaker transformer (M2Former) for far-field multi-speaker ASR in this paper. Experiments on the SMS-WSJ benchmark show that the M2Former outperforms the neural beamformer, MCT, dual-path RNN with transform-average-concatenate and multi-channel deep clustering based end-to-end systems by 9.2%, 14.3%, 24.9%, and 52.2% respectively, in terms of relative word error rate reduction.
title Multi-channel multi-speaker transformer for speech recognition
topic Sound
Artificial Intelligence
url https://arxiv.org/abs/2601.02688