Skip to content
EntityQ30315025· pop 8· linked from 783 articles

Google Wavenet

Sign in to save

Also known as WaveNet, Wavenet

WaveNet is a deep neural network for generating raw audio. It was created by researchers at London-based AI firm DeepMind. The technique, outlined in a paper in September 2016, is able to generate relatively realistic-sounding human-like voices by directly modelling waveforms using a neural network method trained with recordings of real speech. Tests with US English and Mandarin reportedly showed that the system outperforms Google's best existing text-to-speech (TTS) systems, although as of 2016 its text-to-speech synthesis still was less convincing than actual human speech. WaveNet's ability

Wikidata facts

Instance of
generative model

via Wikidata · CC0

~7 min read

Encyclopedic overview

9 sections
Contents
  • History
  • Design and ongoing research
  • Background
  • Initial concept and results
  • Content (voice) swapping
  • Applications
  • See also
  • References
  • External links

WaveNet is a deep neural network for generating raw audio. It was created by researchers at London-based AI firm DeepMind. The technique, outlined in a paper in September 2016, is able to generate relatively realistic-sounding human-like voices by directly modelling waveforms using a neural network method trained with recordings of real speech. Tests with US English and Mandarin reportedly showed that the system outperforms Google's best existing text-to-speech (TTS) systems, although as of 2016 its text-to-speech synthesis still was less convincing than actual human speech. WaveNet's ability to generate raw waveforms means that it can model any kind of audio, including music.

==History==

Excerpted from Wikipedia’s “Google Wavenet” article, available under the CC BY-SA 4.0 licence.

Available in 8 languages

via Wikidata sitelinks · CC0