TTS_tf

Tensorflow implementation of https://github.com/mozilla/TTS

TODO

• Catching up with TTS pytorch

  • Replace pytorch layers with TF
  • Basic eager mode training
    • 5x times slower than pytorch (if you see any easy optimizations, please let me know.)
  • Optimization with tf.function
    • tf.function looks so hard to use with models having different input shapes per iteration since TF retraces the graph for each input. As far as I see from the official examples, workarounds take the advantage of using eager mode. (Correct me if I am wrong.)
  • TF based data loader

• Model exporting pipeline for deployment

• Vocoder adaptation

  • WaveRNN
  • LPCNet