Notes:
WaveNet is a deep learning model that was developed by researchers at DeepMind in 2016. It is designed to generate raw audio waveforms, which can be used to synthesize speech or music.
WaveNet is based on a convolutional neural network architecture, which allows it to process large amounts of data and make predictions based on patterns in the data. It is trained on a dataset of audio waveforms, and is able to generate new waveforms that are similar to the ones in the training data.
One of the key advantages of WaveNet is that it is able to generate audio waveforms at a high sample rate, which allows it to synthesize audio with a high level of detail and naturalness. This has made it a popular choice for tasks such as text-to-speech synthesis and music generation.
Resources:
- lyrebird.ai .. clone your own voice or get a unique custom voice
- modulate.ai .. realistic and fully customizable voice skins
- wellsaidlabs.com .. create life-like synthetic voice for next generation experiences
Wikipedia:
References:
- Tasker 5.7 app update brings WaveNet Voices, Keyboard Action (30 Apr 2019)
- Mammoth Tasker v5.7 update brings WaveNet voices, nav bar actions and more (29 Apr 2019)
- Google Cloud Text-to-Speech adds 31 WaveNet voices, 7 languages and dialects (21 Feb 2019)
- Tasker beta can now use WaveNet tech for natural-sounding speech, including an old lady voice (15 Feb 2019)
- Tasker 5.7.0 Beta 2 now available with WaveNet, App Factory fixed (14 Feb 2019)
- Latest Tasker beta lets you use Google’s natural-sounding WaveNet voices for Text to Speech (13 Feb 2019)
- Voysis Announces Embedded WaveNet for Mobile Devices and IoT (27 Sep 2018)
- Voysis develops offline WaveNet voice synthesis model for mobile devices (27 Sep 2018)
- Google Cloud’s Text-to-Speech API moves to GA, adds new WaveNet voices (28 Aug 2018)
See also:
100 Best DeepMind Videos | 100 Best Speech Synthesis Videos
- Google Text to Speech With WaveNet AI
- Google Text to Speech Without WaveNet AI
- WSLTTS vs WaveNet
- Tasker – Testing WaveNet voices
- Chrome Text to Speech pt2: Wavenet Quality Voices and Speed (10x)
- All Google Wavenet Voices tested
- TECH-UPDATE: DJI Mavic 2 and Google’s text-to-speech progress with WaveNet
- Norman Casagrande: WaveNet, what’s behind Google’s voice
- DeepMind’s WaveNet How it works, and how it is evolving || TensorFlow and Deep Learning
- Your Apps Can Talk! Introducing Cloud Text-to-Speech, Powered by WaveNet Technology (Cloud Next ’18)
- Google Wavenet | Google I/O 2018
- Hans-Hermann Hoppe – A Theory of Socialism and Capitalism (Google WaveNet Voice)
- Hans-Hermann Hoppe – Democracy: The God That Failed – Audiobook (Google WaveNet Voice)
- Google I/O 2018: All the new features coming to the Google Assistant | Digit.in
- Google’s Duplex | Google’s new talking AI just beat the Turing test!
- WaveNet trained on Hide and Seek | Machine Learning
- DeepMind’s WaveNet, 1000 Times Faster | Two Minute Papers #232
- Google’s Text Reader AI: Almost Perfect | Two Minute Papers #228
- DeepMind’s WaveNet : How it works, and how it is evolving – TensorFlow and Deep Learning
- Google’s AI-driven TTS engine
- Deepmind’s wavenet helps make google assistant sound more human
- Generating Putin’s Voice via Neural Networks [WaveNet]
- WaveNet Podcast
- An A.I. creating music by itself – WaveNets
- WaveNet machine learning algorithm speaking
- DeepMind is Improving the Voice of Robots
- rough piano transcription of deepmind’s wavenet piano sample 3 (sageev oore)
- Google DeepMind A.I. Composes and Performs Piano Beyond The Masterpiece Level in Realtime
- WaveNet by Google DeepMind | Two Minute Papers #93
- DeepMind’s WaveNet Text-To-Speech Algorithm
- Google WaveNet: Künstliche Sprache – Neuronales Netz klingt 90% menschlich
- Closed-captioning WaveNets