This New Tech Can Copy Anyone’s Voice Using Just a Minute of Audio

A Canadian AI startup named Lyrebird has launched a new API that allows people to copy anyone's voice using just a one-minute recording of anyone's voice audio.


Scientists always try to create image editing software or better algorithms for visual recognition with new features that work in unique ways. In short, scientists put a lot of effort to improve new technologies. Similarly, new voice-editing tech is on the way. A Canadian AI startup named Lyrebird developed a new tech that can copy anyone’s voice.

The company has unveiled its first product: a set of algorithms that can copy anyone’s voice by listening to just a single minute of sample audio. Through this tech, users could generate or design dialogs/ voice with the voice of their choice for their needs.

Through this invention, Lyrebird is moving forward in the development of AI applications. It takes a bit of computing power to generate a voiceprint.

Previously, by using Artificial Intelligence, the company Adobe had unveiled a project VoCo that can edit human speech like Photoshop tweaks digital images. It requires at least 20 minutes of sample audio before it can mimic a voice. Like VoCo, this latest application program interface (API) synthesizes speech using anyone’s voice. But the difference is, it only needs a minute to synthesize a long sample voice.

Lyrebird said, “its algorithms can also infuse the speech it creates with emotion, letting customers make voices sound angry, sympathetic, or stressed out. The resulting speech can be put to a wide range of uses, including the reading of audio books with famous voices, for connected devices of any kind, for speech synthesis for people with disabilities, for animation movies or for video game studios.”

“By releasing our technology publicly and making it available to anyone, we want to ensure that there will be no such risks.”

Currently, the tech is still under construction. The company has not yet revealed its price. Still, the company said, “more than 6,000 individuals have signed up for early access to its APIs, and we are working to improve its algorithms, including adding support for different languages like French.”

Below you can hear the synthesized voices of Donald Trump, Barack Obama, and Hillary Clinton discussing the startup: