BUT System for the MLC-SLM Challenge

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Polok, Alexander, Han, Jiangyu, Klement, Dominik, Cornell, Samuele, Černocký, Jan, Burget, Lukáš
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916795863007232
author Polok, Alexander
Han, Jiangyu
Klement, Dominik
Cornell, Samuele
Černocký, Jan
Burget, Lukáš
author_facet Polok, Alexander
Han, Jiangyu
Klement, Dominik
Cornell, Samuele
Černocký, Jan
Burget, Lukáš
contents We present a two-speaker automatic speech recognition (ASR) system that combines DiCoW -- a diarization-conditioned variant of Whisper -- with DiariZen, a diarization pipeline built on top of Pyannote. We first evaluate both systems in out-of-domain (OOD) multilingual scenarios without any fine-tuning. In this scenario, DiariZen consistently outperforms the baseline Pyannote diarization model, demonstrating strong generalization. Despite being fine-tuned on English-only data for target-speaker ASR, DiCoW retains solid multilingual performance, indicating that encoder modifications preserve Whisper's multilingual capabilities. We then fine-tune both DiCoW and DiariZen on the MLC-SLM challenge data. The fine-tuned DiariZen continues to outperform the fine-tuned Pyannote baseline, while DiCoW sees further gains from domain adaptation. Our final system achieves a micro-average tcpWER/CER of 16.75% and ranks second in Task 2 of the MLC-SLM challenge. Lastly, we identify several labeling inconsistencies in the training data -- such as missing speech segments and incorrect silence annotations -- which can hinder diarization fine-tuning. We propose simple mitigation strategies to address these issues and improve system robustness.
format Preprint
id arxiv_https___arxiv_org_abs_2506_13414
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BUT System for the MLC-SLM Challenge
Polok, Alexander
Han, Jiangyu
Klement, Dominik
Cornell, Samuele
Černocký, Jan
Burget, Lukáš
Audio and Speech Processing
We present a two-speaker automatic speech recognition (ASR) system that combines DiCoW -- a diarization-conditioned variant of Whisper -- with DiariZen, a diarization pipeline built on top of Pyannote. We first evaluate both systems in out-of-domain (OOD) multilingual scenarios without any fine-tuning. In this scenario, DiariZen consistently outperforms the baseline Pyannote diarization model, demonstrating strong generalization. Despite being fine-tuned on English-only data for target-speaker ASR, DiCoW retains solid multilingual performance, indicating that encoder modifications preserve Whisper's multilingual capabilities. We then fine-tune both DiCoW and DiariZen on the MLC-SLM challenge data. The fine-tuned DiariZen continues to outperform the fine-tuned Pyannote baseline, while DiCoW sees further gains from domain adaptation. Our final system achieves a micro-average tcpWER/CER of 16.75% and ranks second in Task 2 of the MLC-SLM challenge. Lastly, we identify several labeling inconsistencies in the training data -- such as missing speech segments and incorrect silence annotations -- which can hinder diarization fine-tuning. We propose simple mitigation strategies to address these issues and improve system robustness.
title BUT System for the MLC-SLM Challenge
topic Audio and Speech Processing
url https://arxiv.org/abs/2506.13414