Add pretrained models: JSUT and LibriTTS

[r9y9]

JSUT

• Transformer
• Tacotron 2
• MoL WaveNet

Example usage:

cd egs/jsut/tts1
python -c "import pyopenjtalk;print(pyopenjtalk.g2p('こんにちは。私はグーグルアシスタントです。'))" > test.txt
../../../utils/synth_wav.sh --stage 0 --stop-stage 4 --models jsut.transformer.v1 --vocoder-models jsut.wavenet.mol.v1 --trans_type phn --fs 24000 --fmin 80 --fmax 7600 test.txt

LibriTTS

• Universal MoL WaveNet

Example usage:

cd egs/libritts/tts1
echo "This is the demonstration of text-to-speech." > test.txt
../../../utils/synth_wav.sh --stage 0 --stop-stage 4 --models libritts.transformer.v1 --vocoder-models libritts.wavenet.mol.v1 --fs 24000 --input_wav /path/to/libritts/clean_test/237_134493_000003_000004.wav test.txt

Demo samples: https://drive.google.com/open?id=1QOo16aFGa5jN3Sxy1rREU9LS5QFm8jMV

[codecov]

Codecov Report

Merging #1260 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master    #1260   +/-   ##
=======================================
  Coverage   78.08%   78.08%           
=======================================
  Files         102      102           
  Lines        9478     9478           
=======================================
  Hits         7401     7401           
  Misses       2077     2077

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b69306b...522b6c7. Read the comment docs.

[kan-bayashi]

Could you also update README.md?

[r9y9]

README update done

[kan-bayashi]

LGTM!
I will merge after CI.