Knowledge-and-Data-Driven Amplitude Spectrum Prediction for Hierarchical Neural Vocoders
This work addresses speech synthesis quality for text-to-speech applications, representing an incremental improvement over existing neural vocoder methods.
The paper tackles the problem of improving speech synthesis quality in neural vocoders by proposing a knowledge-and-data-driven amplitude spectrum predictor (KDD-ASP) that combines theoretical models with data-driven refinement, resulting in higher-quality synthetic speech compared to conventional methods and WaveRNN on a text-to-speech task.
In our previous work, we have proposed a neural vocoder called HiNet which recovers speech waveforms by predicting amplitude and phase spectra hierarchically from input acoustic features. In HiNet, the amplitude spectrum predictor (ASP) predicts log amplitude spectra (LAS) from input acoustic features. This paper proposes a novel knowledge-and-data-driven ASP (KDD-ASP) to improve the conventional one. First, acoustic features (i.e., F0 and mel-cepstra) pass through a knowledge-driven LAS recovery module to obtain approximate LAS (ALAS). This module is designed based on the combination of STFT and source-filter theory, in which the source part and the filter part are designed based on input F0 and mel-cepstra, respectively. Then, the recovered ALAS are processed by a data-driven LAS refinement module which consists of multiple trainable convolutional layers to get the final LAS. Experimental results show that the HiNet vocoder using KDD-ASP can achieve higher quality of synthetic speech than that using conventional ASP and the WaveRNN vocoder on a text-to-speech (TTS) task.