Improving Code-Switching and Named Entity Recognition in ASR with Speech Editing based Data Augmentation
This addresses the challenge of generating realistic and diverse training data for ASR in specialized scenarios like code-switching and NER, though it is incremental as it builds on existing data augmentation techniques.
The paper tackled the problem of poor performance of end-to-end ASR models in code-switching and named entity recognition by proposing a speech editing-based data augmentation method, which significantly outperformed audio splicing and TTS-based systems in experiments.
Recently, end-to-end (E2E) automatic speech recognition (ASR) models have made great strides and exhibit excellent performance in general speech recognition. However, there remain several challenging scenarios that E2E models are not competent in, such as code-switching and named entity recognition (NER). Data augmentation is a common and effective practice for these two scenarios. However, the current data augmentation methods mainly rely on audio splicing and text-to-speech (TTS) models, which might result in discontinuous, unrealistic, and less diversified speech. To mitigate these potential issues, we propose a novel data augmentation method by applying the text-based speech editing model. The augmented speech from speech editing systems is more coherent and diversified, also more akin to real speech. The experimental results on code-switching and NER tasks show that our proposed method can significantly outperform the audio splicing and neural TTS based data augmentation systems.