Contextual Adapters for Personalized Speech Recognition in Neural Transducers
This addresses the problem of personalization in ASR for users, offering an incremental improvement over existing methods.
The paper tackles the challenge of personal rare word recognition in end-to-end Automatic Speech Recognition (E2E ASR) models by proposing neural contextual adapters for neural transducer-based ASR, which outperforms shallow fusion methods without altering pretrained model weights.
Personal rare word recognition in end-to-end Automatic Speech Recognition (E2E ASR) models is a challenge due to the lack of training data. A standard way to address this issue is with shallow fusion methods at inference time. However, due to their dependence on external language models and the deterministic approach to weight boosting, their performance is limited. In this paper, we propose training neural contextual adapters for personalization in neural transducer based ASR models. Our approach can not only bias towards user-defined words, but also has the flexibility to work with pretrained ASR models. Using an in-house dataset, we demonstrate that contextual adapters can be applied to any general purpose pretrained ASR model to improve personalization. Our method outperforms shallow fusion, while retaining functionality of the pretrained models by not altering any of the model weights. We further show that the adapter style training is superior to full-fine-tuning of the ASR models on datasets with user-defined content.