Categories

LPCNet – Open Source Neural Net speech synthesis

Posted on Friday, November 30th, 2018 in open source by DP

sampling200

Jean-Marc Valin has been working on Neural Network (NN) based speech synthesis in his project called LPCNet:

This new demo presents LPCNet, an architecture that combines signal processing and deep learning to improve the efficiency of neural speech synthesis. Neural speech synthesis models like WaveNet have recently demonstrated impressive speech synthesis quality. Unfortunately, their computational complexity has made them hard to use in real-time, especially on phones. As was the case in the RNNoise project, one solution is to use a combination of deep learning and digital signal processing (DSP) techniques. This demo explains the motivations for LPCNet, shows what it can achieve, and explores its possible applications

Via Rowetel.

This entry was posted on Friday, November 30th, 2018 at 11:21 pm and is filed under open source. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

Leave a Reply

Notify me of followup comments via e-mail. You can also subscribe without commenting.

Recent Comments

  • Joe Desbonnet: Ya, I can recommend the low melting point solder. I used brand 'ChipQuik' and it's amazingly easy to use.
  • Jerome: I need a new BusPirate for the Fablab ;) Many thanks!
  • Max: Seems like an unexpectedly violent way to remove the chip indeed. A hot air station should of course do the job just fine, but in...
  • jose: Part removal described here is pure butchery, the cheapest hot air station will do a fast and clean job removing the QFP, heat air to...
  • Cody: Yes please