Problem
Research questions and friction points this paper is trying to address.
Cross-Lingual ASR
Low-Resource Languages
Parameter-Efficient Transfer
Sequential Adapter Stacking
Sequential Adapter Stacking for Cross-Lingual Low-Resource ASR
🤖 AI Summary
研究通过在资源丰富语言适配器上叠加目标语言适配器的方法，提高低资源语言的自动语音识别性能，实验显示该方法优于完全微调。
📝 Abstract
Extending large-scale multilingual automatic speech recognition (ASR) models to low-resource languages remains challenging. Model performance is skewed toward high-resource languages and degrades sharply for languages with limited labeled data and pre-training exposure. To address this, we investigate parameter-efficient approaches for transferring knowledge from resource-rich source languages to low-resource target languages on Whisper. Alongside warm initialization and attention-based fusion, we propose Sequential Adapter Stacking, which places a trainable target-language adapter on top of a frozen source-language adapter. Under controlled experiments, these approaches are evaluated on three target languages unsupported by Whisper -- Asturian, Assamese, and Xhosa -- using source languages with varying degrees of relatedness. Sequential Adapter Stacking with the closest related source consistently and significantly outperforms full fine-tuning across the three targets, with 5--8\% relative WER reductions. These gains largely persist with only one hour of target training data.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Sequential Adapter Stacking
Cross-Lingual ASR
Low-Resource Languages
Parameter-Efficient Transfer
🔎 Similar Papers
No similar papers found.
