DeepMind publishes WaveNet, a model of raw audio

DeepMind published WaveNet, a generative network that modeled raw audio waveforms directly. DeepMind said it closed more than half the quality gap between existing text-to-speech systems and recorded human speech, and that it also generated music.

Why it mattered The method later ran behind Google’s production text-to-speech, moving neural audio generation from a research demonstration into a shipped service.