Deep learning-based approaches for text to speech application

The proposed research project is aimed at the development of text-to-speech systems that leverage the power of deep neural networks. Our goal is to leverage powerful generated models for speech and combine them with mechanisms to incorporate text information. We hope to explore several such techniques and propose novel methods to achieve this goal.
The development of such systems is essential for allowing us to interact with machines in an intuitive way. The development of natural sounding text-to-speech systems has many potential applications, from conversational agents on phones and mobile devices to educational tools and toys. This technology also has applications in the medical domain, and can be used to help/augment people with speaking disabilities.

Faculty Supervisor:

James Clark

Student:

Partner:

Huawei Technologies Canada Co Ltd (Montreal, QC)

Discipline:

Computer science

Sector:

Technology

University:

McGill University

Program: