TensorX
返回文献探索

Paper · arXiv 2506.13414

BUT System for the MLC-SLM Challenge

Alexander Polok, Jiangyu Han, Dominik Klement, Samuele Cornell, Jan Černocký, Lukáš Burget

15 upvotesJune 16, 2025arXiv 预印本
AI 摘要

The combined DiCoW and DiariZen ASR system demonstrates strong performance in multilingual scenarios, with DiCoW preserving its multilingual capabilities and DiariZen improving through fine-tuning.

DiCoWDiariZenWhisperPyannoteASRout-of-domainmultilingualfine-tuningmicro-average tcpWER/CERMLC-SLM challengelabeling inconsistencies

Abstract

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.

北京市昌平区探索星信息技术及软件开发工作室

京ICP备2026059466号
BUT System for the MLC-SLM Challenge | TensorX