Re:A 2019 Guide to Speech Synthesis with Deep Learning
by PieSquared on Wednesday August 28, 2019 @08:20PM
(#20825473)
I don't think anything about the current set of tools is specific to sample rate; WaveNet, Tacotron, WaveRNN, etc, should work fine to generate 44.1Khz audio. They might just need slightly different hyperparameters or sizes to work well, or may take longer to train due to longer sequence lengths.
If we're stuck with downsampling to 16 Khz, my question still stands.