We are a team of computer science our graduation project about EmotionalSpeech Synthesis.

We've found valuable information like research papers and WaveNet, Tacotron. A website (https://www.voicery.com/)

we were hoping to get to know more information from you.

We need more details what should we start with to grasp the fundamentals to build this idea, what is the architecture to be used in this project, whether there are papers, a GitHub Repository containing helpful documentation, datasets, some other resources, previous knowledge.


1 Answer 1


Since I am working with audio synthesis projects, I can suggest you to go through the recently released paper "MelNet: A Generative Model for Audio in the Frequency Domain", in this architecture human like speech is synthesized, I think this might be better than WaveNet where the generated audio does contain robotic nature, I think this will definitely help you with Emotional speech synthesis as it generates human-like voices. You can check for online demos as well.


Here Is the Paper!

Related YouTube Video

Hope this helps!

  • $\begingroup$ Thanks for helping the demo is really great, If you please send the paper again as the link doesn't work. $\endgroup$ Commented Oct 17, 2019 at 5:44
  • $\begingroup$ Hi @AndrewMaher Glad that it helped you,I have edited the link. If this or any answer has solved your question please consider accepting it by clicking the check-mark so that others will be benefited. $\endgroup$
    – SUMAN
    Commented Oct 17, 2019 at 7:05
  • $\begingroup$ Can you suggest good datasets to start with $\endgroup$ Commented Oct 19, 2019 at 12:25

Not the answer you're looking for? Browse other questions tagged .