redsh

The only positive thing I see about them is that it forces people to read code and get some understanding of the project they’re working on. But in general, we should rebel against code reviews - they are one of the things that make programming a corporate chore and not something fun. 99% of the time is spent on useless nitpicks and the rest of the 1% can be done with more testing or design sessions.

Recent advances in text-to-speech (TTS) synthesis, such as Tacotron and WaveRNN, have made it possible to construct a fully neural network based TTS system, by coupling the two components together. Such a system is conceptually simple as it only takes grapheme or phoneme input, uses Mel-spectrogram as an intermediate feature, and directly generates speech samples. The system achieves quality equal or close to natural speech. However, the high computational cost of the system and issues with robustness have limited their usage in real-world speech synthesis applications and products. In this paper, we present key modeling improvements and optimization strategies that enable deploying these models, not only on GPU servers, but also on mobile devices. The proposed system can generate high-quality 24 kHz speech at 5x faster than real time on server and 3x faster than real time on mobile devices.

Sorry about this. Record as much voice as you can now (stereo too?), then you’ll have time to find the right solution and improve it as the technology gets better in time

I am implementing a 360 panorama generator in my photo effects app http://wonderc.am