Multilingual Voice Recognition

Have you ever imagined a world where you can control everything at home and at work with your voice, even without having to read an instruction manual? Speech technologies based on artificial intelligence models promise to make that happen by teaching machines to recognize voice. However, these systems currently support only a few languages, and it remains a particular challenge to support languages for which large datasets do not exist. The proposed research will explore various state-of-the-art deep learning techniques to improve the performance of voice recognition for low-resource languages. This will help Picovoice to provide the same quality of service for diverse users and help language communities, where language is the only means of expression, to access important information they could not otherwise obtain.

Faculty Supervisor:

Bernd Stelzer

Student:

Partner:

Picovoice

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

Simon Fraser University

Program: