Spotify has introduced a novel artificial intelligence-driven feature designed to translate podcasts into various languages while preserving the host’s authentic voice. This innovative feature relies on OpenAI’s voice generation technology, which has been announced in conjunction with the new offering.
OpenAI’s technology is capable of generating “realistic synthetic voices” based on just a few seconds of recorded speech, making it an ideal tool for Spotify’s language translation feature. By using this technology, Spotify aims to enhance the podcast listening experience by maintaining the distinctive speech characteristics of the original podcasters.
To develop and test this feature, Spotify collaborated with prominent podcasters such as Dax Shepard, Lex Fridman, Monica Padman, Steven Bartlett, and Bill Simmons, who participated in a select number of past and upcoming episodes. Initially, these episodes have been translated into Spanish, French, and German, and they will soon be accessible to both free users and paid subscribers.
Spotify’s choice of these languages is strategic, as they are among the most popular languages for podcast listeners on the platform. The company has begun with Spanish and plans to follow up quickly with German and French translations. While specific pricing details were not disclosed, Spotify is actively exploring opportunities to extend this service to podcasters worldwide.
Spotify’s commitment to this endeavor is evident in its collaboration with comedian Trevor Noah for his upcoming podcast, signaling the company’s dedication to expanding this innovative feature. As Spotify put it, “This is just the beginning,” indicating its continued efforts to enhance the podcasting experience for its global audience.
By Impact Lab