The Future. OpenAI is bringing voice assistant capabilities to ChatGPT, giving the generative AI something of a technological personality in the way we anthropomorphize Siri or Alexa. With Apple and Amazon also working on generative AI tools, expect those voice assistants to get an LLM update… potentially making the use of AI a new normal.
ChatGPT can now answer your queries out loud.
- ChatGPT’s new voice feature was developed using a new text-to-speech model that can create “human-like voices” from “a few seconds of sampled speech.”
- OpenAI also said it teamed up with professional voice actors to create five default voices to choose from.
- The feature will be available as an option for paying Plus and Enterprise subscribers in the next two weeks.
As a vote of confidence, Spotify is also slated as a launch partner, with the streamer announcing a new podcast feature that allows creators to share their voices so they can be translated into different languages in their own voice. That’s pretty wild. But the feature is still in a pilot phase with select creators like Dax Shepard and Monica Padman of Armchair Expert and The Ringer’s Bill Simmons.
But with great power comes great risk. ChatGPT’s voice feature could amp up AI-related scams to a whole new level — a threat OpenAI readily admits.
At the very least, it could make Yahoo’s ChatGPT-powered Fantasy Football draft recaps even more cutting.