多語言獅城系統：基於Qwen3-ASR平衡微調的新加坡高效多語種語音識別

摘要

我们推出Polyglot-Lion系列——专为新加坡多语言场景定制的紧凑型自动语音识别模型，涵盖英语、华语、泰米尔语和马来语。该系列模型通过纯公开语音语料库对Qwen3-ASR-0.6B与Qwen3-ASR-1.7B进行微调，采用均衡采样策略确保各语言训练语句数量均等，并刻意省略语言标签条件机制，使模型能从音频中隐式识别语言。在涵盖四种目标语言的12个基准测试中，Polyglot-Lion-1.7B的平均错误率为14.85%，与体积大6倍的MERaLiON-2-10B-ASR模型（14.32%）性能相当，而训练成本仅需单张RTX PRO 6000 GPU耗费81美元，远低于128张GPU基线模型的18,862美元。其推理吞吐量达每秒0.10个样本，较MERaLiON的2.02秒/样本提升约20倍。这些结果表明，对中等规模预训练模型进行语言均衡微调，能以远低于大型专业系统的成本获得可直接部署的多语言ASR方案。

English

We present Polyglot-Lion, a family of compact multilingual automatic speech recognition (ASR) models tailored for the linguistic landscape of Singapore, covering English, Mandarin, Tamil, and Malay. Our models are obtained by fine-tuning Qwen3-ASR-0.6B and Qwen3-ASR-1.7B exclusively on publicly available speech corpora, using a balanced sampling strategy that equalizes the number of training utterances per language and deliberately omits language-tag conditioning so that the model learns to identify languages implicitly from audio. On 12 benchmarks spanning the four target languages, Polyglot-Lion-1.7B achieves an average error rate of 14.85, competitive with MERaLiON-2-10B-ASR (14.32) - a model 6x larger - while incurring a training cost of \81 on a single RTX PRO 6000 GPU compared to 18,862 for the 128-GPU baseline. Inference throughput is approximately 20x faster than MERaLiON at 0.10 s/sample versus 2.02 s/sample. These results demonstrate that linguistically balanced fine-tuning of moderate-scale pretrained models can yield deployment-ready multilingual ASR at a fraction of the cost of larger specialist systems.

多語言獅城系統：基於Qwen3-ASR平衡微調的新加坡高效多語種語音識別

Polyglot-Lion: Efficient Multilingual ASR for Singapore via Balanced Fine-Tuning of Qwen3-ASR

摘要

Support