SpeakerLM: End-to-End Versatile Speaker Diarization and Recognition with Multimodal Large Language Models
Han Yin, Yafeng Chen, Chong Deng, Luyao Cheng, Hui Wang, Chao-Hong Tan, Qian Chen, Wen Wang, Xiangang Li
TL;DR
SpeakerLM addresses the SDR problem by replacing cascaded SD and ASR pipelines with a unified multimodal LLM that jointly models speech content and speaker attribution. It introduces a flexible speaker registration mechanism and a four-stage training strategy to scale across real and simulated data. Empirical results on Mandarin SDR benchmarks show SpeakerLM outperforms state-of-the-art cascaded baselines on both in-domain and out-of-domain data, and demonstrates robustness to different registration settings and varying numbers of registered speakers. This work demonstrates the potential of multimodal LLMs for scalable, data-driven SDR, enabling more accurate and personalized multi-speaker transcripts.
Abstract
The Speaker Diarization and Recognition (SDR) task aims to predict "who spoke when and what" within an audio clip, which is a crucial task in various real-world multi-speaker scenarios such as meeting transcription and dialogue systems. Existing SDR systems typically adopt a cascaded framework, combining multiple modules such as speaker diarization (SD) and automatic speech recognition (ASR). The cascaded systems suffer from several limitations, such as error propagation, difficulty in handling overlapping speech, and lack of joint optimization for exploring the synergy between SD and ASR tasks. To address these limitations, we introduce SpeakerLM, a unified multimodal large language model for SDR that jointly performs SD and ASR in an end-to-end manner. Moreover, to facilitate diverse real-world scenarios, we incorporate a flexible speaker registration mechanism into SpeakerLM, enabling SDR under different speaker registration settings. SpeakerLM is progressively developed with a multi-stage training strategy on large-scale real data. Extensive experiments show that SpeakerLM demonstrates strong data scaling capability and generalizability, outperforming state-of-the-art cascaded baselines on both in-domain and out-of-domain public SDR benchmarks. Furthermore, experimental results show that the proposed speaker registration mechanism effectively ensures robust SDR performance of SpeakerLM across diverse speaker registration conditions and varying numbers of registered speakers.