Meta Unveils Speech Generation AI: Voicebox
• Meta, the parent company of Facebook and Instagram, has announced a speech-generation AI model called Voicebox.
• The AI model does far more than convert text to speech and can match an audio style based on a sample just two seconds long. It also supports six languages: English, French, German, Spanish, Polish, and Portuguese.
• It can edit existing recordings to remove background noise and create speech that is modeled on diverse speech samples.
What Can Voicebox Do?
Voicebox is an AI model from Meta which can generate speech from text. It can match an audio style based on a sample just two seconds long and convert it into another language given a separate speech sample. Furthermore, it can edit existing recordings to remove background noise or create new ones modeled on diverse speech samples.
Who Can Benefit From Voicebox?
Voicebox is said to have various applications for different users; virtual assistants, non-player characters in the metaverse will be able to have realistic voices with this tool as well as content creators and users with accessibility needs.
The Six Supported Languages
Voicebox currently supports six languages: English, French, German, Spanish, Polish and Portuguese. This allows for a wider range of users worldwide who will be able to take advantage of its features no matter what language they speak in their day-to-day lives.
In conclusion, Meta’s Voicebox offers many features such as generating realistic voices from text inputs as well as editing existing recordings to remove background noises or creating new ones that are modelled off diverse voices. With support for 6 languages including English, French, German Spanish Polish and Portuguese; this tool has potential applications for various users such as virtual assistants or content creators with accessibility needs across the world.