Lyrebird AI

Using artificial intelligence to enable creative expression.
Lyrebird is an AI research division within Descript, building a new generation of tools for media editing and synthesis that make content creation more accessible and expressive.

Our work

Overdub Voice : Create a digital voice that sounds like you from a small audio sample.

female-original.wav
female-synthetic.wav
male-original.wav
male-synthetic.wav

Overdub allows you to replace recorded words and phrases with synthesized speech that's tonally blended with the surrounding audio.


Try a live Overdub demo

Compare Descript against other text to speech services: Amazon | Google
Get started for free →

Research

Ethics

With great innovation comes great responsibility. Lyrebird and Descript believe a person's voice is part of their identity. We pledge to do our part for individuals to retain control of their voice. Read more on our Ethics page.

About Lyrebird

Headquartered in Montreal, the Lyrebird team is the AI research division of Descript, the ultimate receptacle of AI-based media synthesis with a real-world application, developing powerful technologies that make content creation easier and more accessible.

Lyrebird was founded by Alexandre de Brébisson, Kundan Kumar, and Jose Sotelo in 2017 while PhD students at MILA, studying under Yoshua Bengio, who won the Turing Prize in 2019 for his pioneering research into deep learning and neural networks.

If you're interested in joining the Lyrebird team to help build the future of media creation, visit our careers page.