Non-autoregressive Mandarin-English Code-switching Speech Recognition
This addresses the challenge of recognizing speech with frequent language switching for East and Southeast Asian users, but it is incremental as it adapts existing methods to a new domain.
The paper tackled Mandarin-English code-switching speech recognition by applying a non-autoregressive framework with Pinyin output and regularization techniques, achieving exciting results on the SEAME corpus.
Mandarin-English code-switching (CS) is frequently used among East and Southeast Asian people. However, the intra-sentence language switching of the two very different languages makes recognizing CS speech challenging. Meanwhile, the recent successful non-autoregressive (NAR) ASR models remove the need for left-to-right beam decoding in autoregressive (AR) models and achieved outstanding performance and fast inference speed, but it has not been applied to Mandarin-English CS speech recognition. This paper takes advantage of the Mask-CTC NAR ASR framework to tackle the CS speech recognition issue. We further propose to change the Mandarin output target of the encoder to Pinyin for faster encoder training and introduce the Pinyin-to-Mandarin decoder to learn contextualized information. Moreover, we use word embedding label smoothing to regularize the decoder with contextualized information and projection matrix regularization to bridge that gap between the encoder and decoder. We evaluate these methods on the SEAME corpus and achieved exciting results.