It’s been a few days since OpenAI introduced ChatGPT’s new advanced voice mode, and the feedback from the small group of ChatGPT Plus subscribers who got early access has been positive. Various online clips showcase the feature’s ability to sing, imitate accents, correct language pronunciation, and narrate stories.
One notable example is from X user @nickfloats, who asked ChatGPT to “tell me a story as if you’re an airline pilot telling it to passengers on a flight.” The chatbot responded almost instantly, altering the audio to sound like it was coming from an intercom. While it struggled with more complex requests like adding engine sounds, the voice itself was clear and emotive, and ChatGPT handled interruptions well.
In a conversation uploaded to YouTube, ChatGPT mentioned it could handle inputs in “dozens of languages,” although the exact number can vary based on dialects and regional variations. One clip showed the chatbot correcting the pronunciation of French words, providing specific tips on adjusting inflection. Another demo featured ChatGPT speaking Turkish in response to a request for an emotive story. Some Turkish X users noted the accent wasn’t native, but ChatGPT completed the story and reacted appropriately by laughing and crying at certain points.
The bot also did a reasonable job with regional US accents, with one video showcasing examples from New York, Boston, Wisconsin, and a stereotypical “valley girl.” Other videos demonstrated ChatGPT’s singing capabilities, including a blues-style rendition of “Happy Birthday” and humorous attempts to mimic animals like frogs and cats singing the same tune.
The demonstrations featured a few different male and female voices, though they did not include the Scarlett Johansson-like “Sky” voice that was removed in May.
For those feeling left out of these fun demos, OpenAI spokesperson Taya Christianson told The Verge that advanced voice mode would be available to all ChatGPT Plus subscribers (which costs $20 per month) sometime this fall.