Xiaomi-CocktailASR-1 Technical Report

2026-09-10

AI Summary

该研究提出Xiaomi-CocktailASR-1，一种基于大语言模型的端到端目标说话人语音识别架构，通过使用参考语音作为声纹提示直接转录目标说话人的语音，无需语音分离，并具备负样本拒绝能力。

Abstract

Recently, large language model (LLM) based ASR models have achieved significant progress, yet they generally lack support for multi-speaker scenarios, where the cocktail party problem remains a critical bottleneck for further advancing ASR. Existing TS-ASR methods, including end-to-end architectures with speaker embeddings and latest LLM-based explorations suffer from degraded single-speaker performance and the inability to reject when the target speaker is absent. In this paper, we propose Xiaomi-CocktailASR-1, an LLM-based end-to-end TS-ASR architecture. By utilizing reference speech as voiceprint prompts, it directly transcribes the target speaker's speech without requiring speech separation. Xiaomi-CocktailASR-1 maintains competitive performance in single-speaker scenarios, comparable to mainstream ASR models. It also features a negative sample rejection capability, outputting empty text when the target speaker is absent from the mixed speech. Additionally, Xiaomi-CocktailASR-1 supports a Chain-of-Thought (CoT) reasoning mode to provide explicit reasoning steps. Extensive experiments on various synthetic and real-world multispeaker benchmarks demonstrate that Xiaomi-CocktailASR-1 achieves state-of-the-art performance, effectively addressing the cocktail party problem through a unified architecture that balances multispeaker and single-speaker recognition accuracy, along with rejection capability.

Problem

• large language model
• automatic speech recognition
• multi-speaker scenarios
• cocktail party problem
• target speaker rejection

Innovation

• LLM-based
• voiceprint prompts
• negative sample rejection
• Chain-of-Thought (CoT)