CopyNE: Better Contextual ASR by Copying Named Entities
This work addresses the problem of incomplete entity transcriptions in contextual ASR for users relying on accurate named entity recognition, representing an incremental improvement over previous dictionary-based methods.
The paper tackles the challenge of transcribing contextual named entities in automatic speech recognition by introducing CopyNE, a mechanism that copies entities as indivisible wholes from a dictionary, which consistently improves entity transcription accuracy and achieves notable improvements even when based on Whisper.
End-to-end automatic speech recognition (ASR) systems have made significant progress in general scenarios. However, it remains challenging to transcribe contextual named entities (NEs) in the contextual ASR scenario. Previous approaches have attempted to address this by utilizing the NE dictionary. These approaches treat entities as individual tokens and generate them token-by-token, which may result in incomplete transcriptions of entities. In this paper, we treat entities as indivisible wholes and introduce the idea of copying into ASR. We design a systematic mechanism called CopyNE, which can copy entities from the NE dictionary. By copying all tokens of an entity at once, we can reduce errors during entity transcription, ensuring the completeness of the entity. Experiments demonstrate that CopyNE consistently improves the accuracy of transcribing entities compared to previous approaches. Even when based on the strong Whisper, CopyNE still achieves notable improvements.