---
title: "Qwen3-ASR Technical Report"
canonical_url: "https://www.modelscope.cn/papers/237283"
md_url: "https://www.modelscope.cn/papers/237283.md"
arxiv_id: 2601.21337
published: 2026-01-30
last_updated: 2026-01-30
authors:
  - "Xian Shi"
  - "Xiong Wang"
  - "Zhifang Guo"
  - "Yongqi Wang"
  - "Pei Zhang"
  - "Xinyu Zhang"
  - "Zishan Guo"
  - "Hongkun Hao"
  - "Yu Xi"
  - "Baosong Yang"
  - "Jin Xu"
  - "Jingren Zhou"
  - "Junyang Lin"
model_name: Qwen3-ASR
model_developer: "Qwen Team"
domain:
  - "语音识别"
  - "自然语言处理"
  - "多语言处理"
  - "强制对齐"
  - "音频理解"
type:
  - "语音识别"
  - "自然语言处理"
  - "多语言处理"
  - "强制对齐"
  - "音频理解"
  - "Computation and Language"
  - Sound
  - "Audio and Speech Processing"
arxiv_url: "https://arxiv.org/abs/2601.21337"
pdf_url: "https://arxiv.org/pdf/2601.21337"
code_link: "https://github.com/QwenLM/Qwen3-ASR"
---

# Qwen3-ASR Technical Report

> In this report, we introduce Qwen3-ASR family, which includes two powerful all-in-one speech recognition models and a novel non-autoregressive speech forced alignment model. Qwen3-ASR-1.7B and Qwen3-ASR-0.6B are ASR models that support language…

「Qwen3-ASR Technical Report」是 ModelScope 魔搭社区收录的论文，arXiv 2601.21337，作者为 Xian Shi, Xiong Wang, Zhifang Guo et al.，发表于 2026-01-30，属于 语音识别、自然语言处理、多语言处理 领域。

- **ArXiv**: 2601.21337
- **Published**: 2026-01-30
- **Authors**: Xian Shi, Xiong Wang, Zhifang Guo, Yongqi Wang, Pei Zhang, Xinyu Zhang, Zishan Guo, Hongkun Hao, Yu Xi, Baosong Yang, Jin Xu, Jingren Zhou, Junyang Lin
- **Model**: Qwen3-ASR
- **Developer**: Qwen Team
- **Domain**: 语音识别, 自然语言处理, 多语言处理, 强制对齐, 音频理解
- **ArXiv URL**: https://arxiv.org/abs/2601.21337
- **PDF**: https://arxiv.org/pdf/2601.21337
- **Code**: https://github.com/QwenLM/Qwen3-ASR

Source: https://www.modelscope.cn/papers/237283

---

> Qwen3-ASR：面向真实世界的全能型语音识别家族——兼具52语种识别、噪声鲁棒性、毫秒级流式响应与首创LLM驱动多语言强制对齐能力

## 摘要

研究背景：传统端到端ASR模型在噪声鲁棒性、多语言/方言覆盖、长音频理解及时间戳对齐等现实场景中面临瓶颈，而大音频语言模型（LALM）范式凭借其强语言建模与世界知识能力，为解决上述问题提供了新路径。本文提出Qwen3-ASR家族，包含两个全功能ASR模型（Qwen3-ASR-1.7B/0.6B）和一个新型非自回归强制对齐模型（Qwen3-ForcedAligner-0.6B），均基于Qwen3-Omni基础模型构建。方法上，采用AuT音频编码器+Qwen3语言模型的混合架构，支持流式/离线推理；引入多阶段训练策略（AuT预训练→Omni多模态预训练→ASR监督微调→强化学习优化），并设计专用输出格式与上下文感知机制；Qwen3-ForcedAligner首创LLM驱动的多语言、细粒度（词/句/段）时间戳预测。实验表明：1.7B版在开源ASR模型中达到SOTA，媲美顶级商用API；0.6B版实现最优精度-效率平衡（首字节延迟仅92ms，128并发下2000秒语音1秒转录）；ForcedAligner在11种语言上显著超越MFA、NeMo等主流对齐工具，兼具高精度、高效率与强泛化性。贡献在于开源高质量、工业级ASR全栈方案（含权重、推理框架与微调脚本），填补了轻量级、多语言、端到端带时间戳ASR的技术空白。

## Abstract

In this report, we introduce Qwen3-ASR family, which includes two powerful all-in-one speech recognition models and a novel non-autoregressive speech forced alignment model. Qwen3-ASR-1.7B and Qwen3-ASR-0.6B are ASR models that support language identification and ASR for 52 languages and dialects. Both of them leverage large-scale speech training data and the strong audio understanding ability of their foundation model Qwen3-Omni. We conduct comprehensive internal evaluation besides the open-sourced benchmarks as ASR models might differ little on open-sourced benchmark scores but exhibit significant quality differences in real-world scenarios. The experiments reveal that the 1.7B version achieves SOTA performance among open-sourced ASR models and is competitive with the strongest proprietary APIs while the 0.6B version offers the best accuracy-efficiency trade-off. Qwen3-ASR-0.6B can achieve an average TTFT as low as 92ms and transcribe 2000 seconds speech in 1 second at a concurrency of 128. Qwen3-ForcedAligner-0.6B is an LLM based NAR timestamp predictor that is able to align text-speech pairs in 11 languages. Timestamp accuracy experiments show that the proposed model outperforms the three strongest force alignment models and takes more advantages in efficiency and versatility. To further accelerate the community research of ASR and audio understanding, we release these models under the Apache 2.0 license.