[hts-users:04603] Proposed special session on Deep Learning for Speech Synthesis at SLT 2018

["Simon King" <Simon.King@xxxxxxxx>]

We are planning to propose a special session at SLT 2018 (www.slt2018.org) on

Deep Learning for Speech Synthesis

The session will focus on how Deep Neural Networks are rapidly advancing the performance of all areas of TTS, from text processing to waveform generation, using a wide variety of model architectures. We will especially welcome papers that address one of these topics:

• “End-to-end” text-to-speech
• Text processing
• Waveform generation
• Adaptation to speaker, style, etc
• Use of data not specially collected for speech synthesis (‘found’, user-contributed, etc)

In order to get our proposal accepted by the SLT organisers, we need to demonstrate interest. So, if you might submit a paper to this session, please let us know by replying to this email (please send only to Simon.King@xxxxxxxx, not the list) telling us

• your name(s)
• your group / company / university
• how many papers you might submit

Deadline for your reply: Thursday 7th June 2018

The paper submission deadline would be 2nd July 2018

many thanks,
Simon King
Pierre Lanchantin
Tomoki Toda

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