[Subject Prev][Subject Next][Thread Prev][Thread Next][Date Index][Thread Index]

[hts-users:04418] Merlin - a new Neural Network Speech Synthesis System from the University of Edinburgh


We are pleased to announce the release of 

Merlin

a new Neural Network Speech Synthesis System from the Centre for Speech Technology Research, University of Edinburgh

Features:

* Python + Theano
* free software (Apache License Version 2.0)
* can be combined with with your favourite front-end text processor (e.g., Festival) and vocoder (e.g., STRAIGHT, or WORLD)
* layers can be feedforward, mixture density, recurrent, long short-term memory (LSTM), ….
* comes with recipes (in the spirit of Kaldi)

Get it from https://github.com/CSTR-Edinburgh/merlin

At SSW, we’ll be presenting Merlin in the demo session, and in this paper:

Merlin: An Open Source Neural Network Speech Synthesis System
Zhizheng Wu, Oliver Watts, Simon King
(Poster Session 2 - Wednesday, September 14)


See you there!

Zhizheng Wu, Oliver Watts, Simon King
(+ special thanks to Srikanth Ronanki for pre-release testing and improvements)



-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.