2 ms·
This is amazing! I've been using LJS trained models and then cross-training them to target speakers, but this looks like it may produce even higher quality resu
by echelon 8y ago
This is amazing! I've been using LJS trained models and then cross-training them to target speakers, but this looks like it may produce even higher quality results.
I've previously implemented concatenative TTS using unit selection [1]. The quality is spotty, so I'm throwing it out and going with the ML approach, which produces higher fidelity voices even in my own experiments.
My next steps are taking an end-to-end synthesis model and porting it to run cheaply on the CPU.
Thanks so much for making this data available, Mozilla! You're helping democratize this technology for individual engineers and researchers that don't have Google's resources.
[1] http://trumped.com http://trumped.com