Polyglot-Lion：通过平衡微调实现新加坡多语言语音识别的高效模型
Polyglot-Lion: Efficient Multilingual ASR for Singapore via Balanced Fine-Tuning of Qwen3-ASR
AI总结：
本文提出Polyglot-Lion，通过平衡微调Qwen3-ASR模型，在新加坡四种语言上实现高效多语言语音识别，模型在12个基准测试中达到14.85的平均错误率，成本仅为128GPU基线的1/200。
摘要：
我们提出了Polyglot-Lion，一种针对新加坡语言景观定制的紧凑多语言自动语音识别（ASR）模型，涵盖英语、普通话、泰米尔语和马来语。我们的模型通过仅在公开可用的语音语料库上微调Qwen3-ASR-0.6B和Qwen3-ASR-1.7B，使用平衡采样策略，使每种语言的训练语句数量相等，并有意排除语言标签条件，使模型从音频中隐式学习语言。在覆盖四种目标语言的12个基准测试中，Polyglot-Lion-1.7B实现了14.85的平均错误率，与MERaLiON-2-10B-ASR（14.32）相当，后者模型大6倍，但训练成本仅为单块RTX PRO 6000 GPU的81美元，而基线128GPU成本为18,862美元。推理吞吐量是MERaLiON的约20倍，0.10秒/样本 vs 2.02秒/样本。这些结果表明，对中等规模预训练模型进行语言平衡微调可以以远低于更大专用系统的成本实现部署级多语言ASR。
英文摘要：
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.
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。
