Whisper is a multitask and multilingual speech model covering 99 languages. It yields commendable automatic speech recognition (ASR) results in a subset of its covered languages, but the model still underperforms on a non-negligible number of under-represented languages, a problem exacerbated in smaller model versions. In this work, we examine its limitations, demonstrating the presence of speaker-related (gender, age) and model-related (resourcefulness and model size) bias. Despite that, we show that only model-related bias are amplified by quantization, impacting more low-resource languages and smaller models. Searching for a better compression approach, we propose DistilWhisper, an approach that is able to bridge the performance gap in ASR for these languages while retaining the advantages of multitask and multilingual capabilities. Our approach involves two key strategies: lightweight modular ASR fine-tuning of whisper-small using language-specific experts, and knowledge distillation from whisper-large-v2. This dual approach allows us to effectively boost ASR performance while keeping the robustness inherited from the multitask and multilingual pre-training. Results demonstrate that our approach is more effective than standard fine-tuning or LoRA adapters, boosting performance in the targeted languages for both in- and out-of-domain test sets, while introducing only a negligible parameter overhead at inference.

翻译：Whisper是一个覆盖99种语言的多任务多语言语音模型。该模型在其覆盖的部分语言中取得了值得称赞的自动语音识别（ASR）结果，但在数量不可忽视的低资源语言上表现仍不理想，这一问题在较小模型版本中尤为突出。本研究首先考察其局限性，证实了说话人相关（性别、年龄）和模型相关（资源丰富度与模型大小）偏差的存在。尽管如此，我们证明只有模型相关偏差会被量化（quantization）放大，进而对低资源语言和较小模型造成更大影响。为寻求更优的压缩方法，我们提出DistilWhisper——一种能够在保留多任务多语言能力的优势的同时，缩小这些语言ASR性能差距的方案。该方法包含两项关键策略：使用语言特定专家对whisper-small进行轻量级模块化ASR微调，以及通过whisper-large-v2进行知识蒸馏。这种双管齐下的策略使我们能够有效提升ASR性能，同时保持从多任务多语言预训练中继承的鲁棒性。结果表明，我们的方法比标准微调或LoRA适配器更有效，在域内和域外测试集中均提升了目标语言的性能，且推理时仅引入可忽略的参数开销。