@inproceedings{zheng-etal-2025-dncasr,
title = "{DNCASR}: End-to-End Training for Speaker-Attributed {ASR}",
author = "Zheng, Xianrui and
Zhang, Chao and
Woodland, Phil",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-long.899/",
doi = "10.18653/v1/2025.acl-long.899",
pages = "18369--18383",
ISBN = "979-8-89176-251-0",
abstract = "This paper introduces DNCASR, a novel end-to-end trainable system designed for joint neural speaker clustering and automatic speech recognition (ASR), enabling speaker-attributed transcription of long multi-party meetings. DNCASR uses two separate encoders to independently encode global speaker characteristics and local waveform information, along with two linked decoders to generate speaker-attributed transcriptions. The use of linked decoders allows the entire system to be jointly trained under a unified loss function. By employing a serialised training approach, DNCASR effectively addresses overlapping speech in real-world meetings, where the link improves the prediction of speaker indices in overlapping segments. Experiments on the AMI-MDM meeting corpus demonstrate that the jointly trained DNCASR outperforms a parallel system that does not have links between the speaker and ASR decoders. Using cpWER to measure the speaker-attributed word error rate, DNCASR achieves a 9.0{\%} relative reduction on the AMI-MDM Eval set."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zheng-etal-2025-dncasr">
<titleInfo>
<title>DNCASR: End-to-End Training for Speaker-Attributed ASR</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xianrui</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chao</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Phil</namePart>
<namePart type="family">Woodland</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Nabende</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-251-0</identifier>
</relatedItem>
<abstract>This paper introduces DNCASR, a novel end-to-end trainable system designed for joint neural speaker clustering and automatic speech recognition (ASR), enabling speaker-attributed transcription of long multi-party meetings. DNCASR uses two separate encoders to independently encode global speaker characteristics and local waveform information, along with two linked decoders to generate speaker-attributed transcriptions. The use of linked decoders allows the entire system to be jointly trained under a unified loss function. By employing a serialised training approach, DNCASR effectively addresses overlapping speech in real-world meetings, where the link improves the prediction of speaker indices in overlapping segments. Experiments on the AMI-MDM meeting corpus demonstrate that the jointly trained DNCASR outperforms a parallel system that does not have links between the speaker and ASR decoders. Using cpWER to measure the speaker-attributed word error rate, DNCASR achieves a 9.0% relative reduction on the AMI-MDM Eval set.</abstract>
<identifier type="citekey">zheng-etal-2025-dncasr</identifier>
<identifier type="doi">10.18653/v1/2025.acl-long.899</identifier>
<location>
<url>https://aclanthology.org/2025.acl-long.899/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>18369</start>
<end>18383</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T DNCASR: End-to-End Training for Speaker-Attributed ASR
%A Zheng, Xianrui
%A Zhang, Chao
%A Woodland, Phil
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-251-0
%F zheng-etal-2025-dncasr
%X This paper introduces DNCASR, a novel end-to-end trainable system designed for joint neural speaker clustering and automatic speech recognition (ASR), enabling speaker-attributed transcription of long multi-party meetings. DNCASR uses two separate encoders to independently encode global speaker characteristics and local waveform information, along with two linked decoders to generate speaker-attributed transcriptions. The use of linked decoders allows the entire system to be jointly trained under a unified loss function. By employing a serialised training approach, DNCASR effectively addresses overlapping speech in real-world meetings, where the link improves the prediction of speaker indices in overlapping segments. Experiments on the AMI-MDM meeting corpus demonstrate that the jointly trained DNCASR outperforms a parallel system that does not have links between the speaker and ASR decoders. Using cpWER to measure the speaker-attributed word error rate, DNCASR achieves a 9.0% relative reduction on the AMI-MDM Eval set.
%R 10.18653/v1/2025.acl-long.899
%U https://aclanthology.org/2025.acl-long.899/
%U https://doi.org/10.18653/v1/2025.acl-long.899
%P 18369-18383
Markdown (Informal)
[DNCASR: End-to-End Training for Speaker-Attributed ASR](https://aclanthology.org/2025.acl-long.899/) (Zheng et al., ACL 2025)
ACL
- Xianrui Zheng, Chao Zhang, and Phil Woodland. 2025. DNCASR: End-to-End Training for Speaker-Attributed ASR. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 18369–18383, Vienna, Austria. Association for Computational Linguistics.