High quality, lightweight and adaptable TTS using LPCNet

We present a lightweight adaptable neural TTS system with high quality output. The system is composed of three separate neural network blocks: prosody prediction, acoustic feature prediction and Linear Prediction Coding Net as a neural vocoder. This system can synthesize speech with close to natural quality while running 3 times faster than real-time on a standard CPU.

The modular setup of the system allows for simple adaptation to new voices with a small amount of data. Read More

#nlp, #voice

How Europe’s AI ecosystem could catch up with China and the U.S.

From Alibaba and Baidu to Google, Facebook, and Microsoft, China and the United States produced virtually every one of the top consumer AI companies in the world today. That leaves Europe trailing behind the U.S. and China, even though Europe still has the largest community of cited AI researchers.

Startup founders, analysts, and organizations seeking to bring ecosystems together for collective action pondered how the European AI ecosystem can catch up with China and the United States at TechBBQ, a gathering of hundreds of Nordic tech startups held recently in Copenhagen. Read More

#china-vs-us