Ai Dreams Forum
Software & Hardware => Sound Software => Topic started by: Freddy on June 03, 2018, 07:59:39 pm
-
Yet more interesting stuff from Siraj Raval.
Really looking forward to seeing this develop for gaming platforms.
https://www.youtube.com/watch?v=6KHSPiYlZ-U
-
I was intrigued by the video and did look over the github (https://github.com/llSourcell/Neural_Network_Voices) for the code, it hasn't been updated for 2 years now. Has anyone tried it?
I also found the white paper for it here (https://arxiv.org/pdf/1702.07825.pdf).
I found this article (https://lionbridge.ai/articles/how-to-create-your-own-synthetic-voice-with-just-one-hour-of-speech-lyrebird-review/) about a startup that can synthesize celebrity voices where they can get any public figure and put words in their mouth without having to use a human actor.
There is also Voysis wavent (https://www.forbes.com/sites/parmyolson/2017/11/03/this-startups-artificial-voice-sounds-almost-indistinguishable-from-a-humans/#76cc81ab388c) which was acquired by Apple (https://voicebot.ai/2020/04/06/apple-acquires-voice-ai-startup-voysis/)
Anyone on this board have any experience with this kind of Deep Learning?