Reducing Spelling Inconsistencies in Code-Switching ASR using Contextualized CTC Loss
This addresses spelling errors in ASR for multilingual speakers, but it is incremental as it builds on existing CTC-based methods.
The paper tackled the problem of spelling inconsistencies in code-switching automatic speech recognition (ASR) due to phoneme duplication in character-based models, and proposed a Contextualized CTC loss that improved ASR performance on both code-switching and monolingual corpora compared to regular CTC loss.
Code-Switching (CS) remains a challenge for Automatic Speech Recognition (ASR), especially character-based models. With the combined choice of characters from multiple languages, the outcome from character-based models suffers from phoneme duplication, resulting in language-inconsistent spellings. We propose Contextualized Connectionist Temporal Classification (CCTC) loss to encourage spelling consistencies of a character-based non-autoregressive ASR which allows for faster inference. The CCTC loss conditions the main prediction on the predicted contexts to ensure language consistency in the spellings. In contrast to existing CTC-based approaches, CCTC loss does not require frame-level alignments, since the context ground truth is obtained from the model's estimated path. Compared to the same model trained with regular CTC loss, our method consistently improved the ASR performance on both CS and monolingual corpora.