The Tacotron 2 and WaveGlow model form a text-to-speech system that enables user to synthesise a natural sounding speech from raw transcripts.
The Tacotron 2 model is a recurrent sequence-to-sequence model with attention that predicts mel-spectrograms from text. The encoder (blue blocks in the figure below) transforms the whole text into a fixed-size hidden feature representation. This feature representation is then consumed by the autoregressive decoder (orange blocks) that produces one spectrogram frame at a time. In our implementation, the autoregressive WaveNet (green block) is replaced by the flow-based generative WaveGlow.
Figure 1. Architecture of the Tacotron 2 model. Taken from the Tacotron 2 paper.
The WaveGlow model is a flow-based generative model that generates audio samples from Gaussian distribution using mel-spectrogram conditioning (Figure 2). During training, the model learns to transform the dataset distribution into spherical Gaussian distribution through a series of flows. One step of a flow consists of an invertible convolution, followed by a modified WaveNet architecture that serves as an affine coupling layer. During inference, the network is inverted and audio samples are generated from the Gaussian distribution. Our implementation uses 512 residual channels in the coupling layer.
Figure 2. Architecture of the WaveGlow model. Taken from the WaveGlow paper.
The following datasets were used to train this model:
- LJSpeech-1.1 - Dataset consisting of 13,100 short audio clips of a single speaker reading passages from 7 non-fiction books. A transcription is provided for each clip. Clips vary in length from 1 to 10 seconds and have a total length of approximately 24 hours.
Performance numbers for this model are available in NGC.