Efficient Multilingual ASR Finetuning via LoRA Language Experts
This work addresses the curse of multilinguality in ASR for users needing efficient customization, though it is incremental as it builds on existing Whisper and LoRA methods.
The paper tackled the problem of language interference in multilingual automatic speech recognition (ASR) by proposing an efficient finetuning framework using LoRA language experts based on Whisper, achieving approximately 10% and 15% relative performance gains in language-aware and language-agnostic scenarios.
Recent advancements in deep learning have significantly enhanced multilingual automatic speech recognition (ASR) due to the development of advanced model architectures and available large-scale multilingual datasets. Despite that, multilingual ASR still suffers from the curse of multilinguality in that different languages tend to interfere with each other, making it difficult for the ASR model to identify multiple languages effectively while sharing model capacity across them. This paper proposes an efficient finetuning framework for customized multilingual ASR via prepared LoRA language experts based on Whisper. Through LoRA expert fusion or knowledge distillation, our approach achieves better recognition performance on target languages than standard fine-tuning methods. Experimental results demonstrate that the proposed models yield approximately 10\% and 15\% relative performance gains in language-aware and language-agnostic scenarios, respectively.