SpeakerLM: End-to-End Versatile Speaker Diarization and Recognition with Multimodal Large Language Models

TL;DR

SpeakerLM is a unified multimodal large language model that jointly performs speaker diarization and recognition end-to-end, overcoming limitations of cascaded systems and demonstrating superior performance and robustness across diverse scenarios.

Contribution

The paper introduces SpeakerLM, a novel end-to-end multimodal large language model for SDR with a flexible speaker registration mechanism and multi-stage training, advancing beyond traditional cascaded approaches.

Findings

Outperforms state-of-the-art cascaded baselines on public SDR benchmarks.

Demonstrates strong data scaling capability and generalizability.

Ensures robust SDR performance across diverse registration conditions.

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.