The DKU System for Multi-Speaker Automatic Speech Recognition in MLC-SLM Challenge
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912479190188032 |
|---|---|
| author | Lin, Yuke Cheng, Ming Li, Ze Li, Ming |
| author_facet | Lin, Yuke Cheng, Ming Li, Ze Li, Ming |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_09499 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | The DKU System for Multi-Speaker Automatic Speech Recognition in MLC-SLM Challenge Lin, Yuke Cheng, Ming Li, Ze Li, Ming Audio and Speech Processing Sound 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. |
| title | The DKU System for Multi-Speaker Automatic Speech Recognition in MLC-SLM Challenge |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2507.09499 |