SoulX-Transcriber：一个鲁棒的多说话人语音转录端到端框架
SoulX-Transcriber: A Robust End-to-End Framework for Multi-Speaker Speech Transcription
AI总结：
提出SoulX-Transcriber，一个基于LLM的统一多说话人转录系统，通过两阶段训练策略联合建模说话人日志和ASR，提升复杂场景下的说话人区分和转录鲁棒性。
摘要：
自动语音识别（ASR）和大语言模型（LLM）的最新进展显著提升了语音理解能力。然而，多说话人语音转录仍然是一个具有挑战性的任务，受限于高度相似的说话人声音、快速的说话人转换、重叠话语以及不准确的说话人边界分割。这些挑战在真实对话音频中尤为突出，因为说话人动态和声学条件高度可变。本技术报告介绍了SoulX-Transcriber，一个统一的多说话人转录系统，它在基于LLM的框架中联合建模说话人日志（SD）和ASR。SoulX-Transcriber采用两阶段训练策略来提升说话人区分和转录鲁棒性。在第一阶段，说话人感知的多任务连续预训练增强了说话人表示学习和边界感知。在第二阶段，监督微调进一步优化模型，使其在复杂多说话人条件下实现准确的端到端说话人属性转录。SoulX-Transcriber在多个公开基准测试（包括AliMeeting、AISHELL-4和AMI）中展现出强大的性能和鲁棒性，同时保持对多领域场景的高适应性。
英文摘要：
Recent advances in Automatic Speech Recognition (ASR) and Large Language Models (LLMs) have significantly improved speech understanding capabilities. However, multi-speaker speech transcription remains challenging task, constrained by highly similar speaker voices, rapid turn-taking transitions, overlapping utterances and inaccurate speaker boundary segmentation. These challenges become particularly pronounced in real-world conversational audio, where speaker dynamics and acoustic conditions are highly variable. This technical report presents SoulX-Transcriber, a unified multi-speaker transcription system that jointly models speaker diarization (SD) and ASR within an LLM-based framework. SoulX-Transcriber adopts a two-stage training strategy to improve both speaker discrimination and transcription robustness. In the first stage, speaker-aware multi-task continuous pre-training enhances speaker representation learning and boundary perception. In the second stage, supervised fine-tuning further optimizes the model for accurate end-to-end speaker-attributed transcription under complex multi-speaker conditions. SoulX-Transcriber delivers strong performance and robustness across multiple public benchmarks, including AliMeeting, AISHELL-4, and AMI, while maintaining high adaptability to multi-domain scenarios.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
