Spotify’s audio translation feature leverages OpenAI’s emerging audio rendering technology.
Spotify, one of the world’s most popular podcast platforms, is testing the ability to use artificial intelligence to dub podcasts in different languages with the same tone of voice. With this feature, users will now be able to listen to podcasts in different languages in their own language and in the voice of the speaker.
Using OpenAI’s voice generation technology
This feature, developed by Spotify, makes use of OpenAI’s newly released voice generation technology. This technology converts text into sounds, allowing texts in different languages to be dubbed with the same tone of voice.
Currently being tested in three different languages
The test phase took place on the programs of podcasters Monica Padman, Lex Fridman, Bill Simmons and Steven Bartlett. The voice translation feature was tested in Spanish, French and German, with successful results.
Access to users around the world
“Voice Translation matches the creator’s own voice, giving listeners around the world the power to discover and be inspired by new podcasters in a more authentic way than ever before,” said Ziad Sultan, Spotify’s Vice President of Personalization. We believe a thoughtful approach to AI can help create deeper connections between listeners and creators, which is a key component of Spotify’s mission to unlock the potential of human creativity.”
Over 100 million podcast listeners
Spotify stated that it currently has more than 5 million podcast titles available in more than 170 markets and has more than 100 million podcast listeners worldwide.
Coming to more languages and creators
Spotify has stated that it wants to include more creators in the feature and will continue to develop it in different languages.
How does voice translation work?
Spotify’s voice translation feature makes use of OpenAI’s newly released voice generation technology. This technology converts text into sounds, allowing texts in different languages to be dubbed with the same tone of voice.
The process works like this:
First, the text of the podcast in its original language is passed to an AI.
The AI translates the text into the target language.
The AI voices the text in the target language.
The audio is edited to sound similar to the podcast’s original tone of voice.
How does it contribute to Spotify’s mission?
Spotify is a company with a mission to “bring inspiring voices to people around the world”. The audio translation feature contributes to this mission in two ways:
It makes podcasts in different languages accessible to a wider audience.
It gives podcast listeners access to content from different cultures and perspectives.
