Fetching the paper…

Neural source-filter waveform models for statistical parametric speech synthesis · Around