Towards End-to-end Automatic Code-Switching Speech Recognition
This work addresses speech recognition for mixed-language speakers, but it is incremental as it builds on existing CTC and language model techniques.
The paper tackles the problem of end-to-end automatic speech recognition for English-Mandarin code-switching by proposing a CTC-based model trained on monolingual datasets and fine-tuned with mixed-language data, resulting in a 5% improvement in character error rate (CER).
Speech recognition in mixed language has difficulties to adapt end-to-end framework due to the lack of data and overlapping phone sets, for example in words such as "one" in English and "wàn" in Chinese. We propose a CTC-based end-to-end automatic speech recognition model for intra-sentential English-Mandarin code-switching. The model is trained by joint training on monolingual datasets, and fine-tuning with the mixed-language corpus. During the decoding process, we apply a beam search and combine CTC predictions and language model score. The proposed method is effective in leveraging monolingual corpus and detecting language transitions and it improves the CER by 5%.