coqui-ai / tts-papers Goto Github PK

Hello Coqui team,

Thank you for collecting many related works.

I saw the tutorial "End-to-End Text-to-Speech Synthesis, Part 1 ". I am the presenter : ).
That talk was quite old. The contents were based on my understanding by then and may be inaccurate or confusing.

Therefore, if it is possible, could I recommend some replacement?
(the slides are on dropbox, and the link address may be too long to show here. Thus, I showed my github page link, from which it is easy to get the slides)

1. On neural vocoder, I and Prof. Yamagishi made a few tutorials on neural vocoders, and the latest one was here http://tonywangx.github.io/slide.html#may. The Pytorch-based hands-on materials cover a few presentative neural vocoders, and they are easy to run on the google colab platform. I think they may be useful to the newcomers;

2. On acoustic models, this tutorial given at ISCA Speaker Odyssey 2020 covered a few representative models. They may be old, but I think there is limited work covering both soft- and hard-attention-based approaches. They can be useful too: http://tonywangx.github.io/slide.html#label-slide-2020-nov-2. The videos is here: https://youtu.be/WCe7SYcDzAI.

Finally, Dr. Tan Xu's survey paper provides a comprehensive overview https://arxiv.org/abs/2106.15561.

Thank you!

Unfortunately we, while at Mozilla, forgot to add a license file to this repo when we created it.

As we were at Mozilla at the time we, and Mozilla, intended to use the default license for Mozilla projects, the Mozilla Public License 2.0. I'm tagging contributors to this repo asking you to confirm that you agree to license your contributions as MPL-2.0.

If you agree, please reply with a comment here saying "I agree to license my contributions to this repository under the Mozilla Public License 2.0."

• @octalpixel
• @shaun95