Did your favorite podcaster learn to speak Spanish overnight? No, you’re actually listening to synthetic speech powered by artificial intelligence.
Spotify is using voice-cloning technology to let you hear podcasts in several languages, including Spanish, French, and German. It not only translates the podcaster but does so in their voice. In a video demo, Spotify says a "podcast episode originally recorded in English can now be available in other languages while keeping the speaker’s distinctive speech characteristics."
Spotify didn’t delve into the technical details behind the voice translation or if it had certain limitations. But it looks like the system harnesses the power of voice cloning, which has grown in popularity in recent years. The technology works by taking samples of your voice; a computer model is then trained to generate speech in your voice based on whatever text input it applied.
The technology has raised flags in the security community for its potential to fuel scams. But voice cloning could usher in a new age of real-time translation, which Spotify is seeking to unlock.
“By matching the creator’s own voice, Voice Translation gives listeners around the world the power to discover and be inspired by new podcasters in a more authentic way than ever before,” says Spotify VP of Personalization Ziad Sultan.
For now, Spotify says the company’s “Voice Translation for podcasts” is in pilot mode for a limited number of podcasts from personalities including Dax Shepard, Monica Padman, Lex Fridman, Bill Simmons, and Steven Bartlett for both Spotify Premium and free users. The goal is to expand the technology to more podcasters and additional languages.
"We’ll start rolling these out to users on the Now Playing View of supported episodes starting today,” the company says.