We present the DKU system for Task 2 of the MLC-SLM Challenge, which aims to perform multi-speaker automatic speech recognition directly from raw audio without Oracle speaker labels or time boundaries. Our approach builds upon a diarization-aware framework integrating speaker embeddings and temporal utterance boundaries into a Qwen2.5-based large language model (LLM). Then, we enhance the system's multilingual performance by fine-tuning language-specific adapters and LoRA modules within the LLM decoder. Finally, our system achieves the tcpWER of 23.56\% and 18.08\% on the development and test sets of the MLC-SLM dataset, substantially outperforming the official baseline.
@article{arxiv.2507.09499,
title = {The DKU System for Multi-Speaker Automatic Speech Recognition in MLC-SLM Challenge},
author = {Yuke Lin and Ming Cheng and Ze Li and Ming Li},
journal= {arXiv preprint arXiv:2507.09499},
year = {2025}
}
Comments
Technical Report for MLC-SLM Challenge in Interspeech2025