5 days left to enroll, learn more and signup here -
https://www.theschool.ai/courses/decentralized-applications
Baidu has released some really impressive research that enables them to generate a voice in the style of anyone after having been trained on only a few examples. Few-shot generative learning is something i'm particularly interested in, and in this video I'll go over what their progress has looked like in this field over the past 2 years. We'll go over a web demo of audio generation, try and understand how DeepMind's WaveNet (similar) works, and then look at some Tensorflow code to get a deeper understanding of how this model plays out programmatically.
Code for this video:
https://github.com/llSourcell/Neural_Network_Voices
Please Subscribe! And like. And comment. That's what keeps me going.
Want more education? Connect with me here:
Twitter: https://twitter.com/sirajraval
Facebook: https://www.facebook.com/sirajology
instagram: https://www.instagram.com/sirajraval
More learning resources:
https://github.com/baidu-research/deep-voice
https://thenextweb.com/artificial-intelligence/2018/02/26/baidus-ai-can-clone-your-voice-and-give-it-a-different-gender-or-accent/
http://research.baidu.com/deep-voice-3-2000-speaker-neural-text-speech/
http://research.baidu.com/neural-voice-cloning-samples/
Join us in the Wizards Slack channel:
http://wizards.herokuapp.com/
And please support me on Patreon:
https://www.patreon.com/user?u=3191693
