Disambiguation of Chinese Polyphones in an End-to-End Framework with Semantic Features Extracted by Pre-trained BERT
This work addresses a core issue in Chinese TTS systems, but it is incremental as it builds on existing BERT and neural network methods.
The paper tackles polyphone disambiguation in Chinese Mandarin text-to-speech by proposing an end-to-end framework using pre-trained BERT for semantic feature extraction and neural network classifiers, achieving enhanced performance compared to an LSTM baseline.
Grapheme-to-phoneme (G2P) conversion serves as an essential component in Chinese Mandarin text-to-speech (TTS) system, where polyphone disambiguation is the core issue. In this paper, we propose an end-to-end framework to predict the pronunciation of a polyphonic character, which accepts sentence containing polyphonic character as input in the form of Chinese character sequence without the necessity of any preprocessing. The proposed method consists of a pre-trained bidirectional encoder representations from Transformers (BERT) model and a neural network (NN) based classifier. The pre-trained BERT model extracts semantic features from a raw Chinese character sequence and the NN based classifier predicts the polyphonic character's pronunciation according to BERT output. In out experiments, we implemented three classifiers, a fully-connected network based classifier, a long short-term memory (LSTM) network based classifier and a Transformer block based classifier. The experimental results compared with the baseline approach based on LSTM demonstrate that, the pre-trained model extracts effective semantic features, which greatly enhances the performance of polyphone disambiguation. In addition, we also explored the impact of contextual information on polyphone disambiguation.