Multilingual Contextual Adapters To Improve Custom Word Recognition In Low-resource Languages
This addresses personalization challenges in ASR for low-resource languages, representing an incremental improvement over existing contextual adapter methods.
The paper tackles the problem of custom word recognition in low-resource languages using CTC-based ASR models, achieving a 48% F1 improvement in retrieving unseen custom entities and a 5-11% WER reduction in the base model.
Connectionist Temporal Classification (CTC) models are popular for their balance between speed and performance for Automatic Speech Recognition (ASR). However, these CTC models still struggle in other areas, such as personalization towards custom words. A recent approach explores Contextual Adapters, wherein an attention-based biasing model for CTC is used to improve the recognition of custom entities. While this approach works well with enough data, we showcase that it isn't an effective strategy for low-resource languages. In this work, we propose a supervision loss for smoother training of the Contextual Adapters. Further, we explore a multilingual strategy to improve performance with limited training data. Our method achieves 48% F1 improvement in retrieving unseen custom entities for a low-resource language. Interestingly, as a by-product of training the Contextual Adapters, we see a 5-11% Word Error Rate (WER) reduction in the performance of the base CTC model as well.