End-to-End Mandarin Tone Classification with Short Term Con Information
In this paper, we propose an end-to-end Mandarin tone classification method from continuous speech utterances utilizing both the spectrogram and the short-term con in-formation as the input. Both spectrograms and con segment features are used to train the tone classifier. We first divide the spectrogram frames into syllable segments using force alignment results produced by an ASR model. Then we extract the short-term segment features to capture the con information across multiple syllables. Feeding both the spectrogram and the short-term con segment features into an end-to-end model could sig-nificantly improve the performance. Experiments are performed on a large-scale open-source Mandarin speech dataset to evaluate the proposed method. Results show that this method improves the classification accuracy from 79.5% to 92.6% on the AISHELL3 database.
