OpenAI recently launched ChatGPT's advanced voice mode, receiving positive feedback from the select group of ChatGPT Plus subscribers initially granted access. Users have shared various clips showcasing the feature's capabilities, such as singing, imitating accents, correcting pronunciation, and engaging in narrative storytelling.
In one example, a user requested ChatGPT to "tell me a story as if you're an airline pilot addressing passengers." The chatbot responded promptly, modifying the audio to mimic an intercom environment. While it faced challenges with more intricate tasks like adding engine sounds, the voice remained clear and expressive, and the chatbot managed interruptions effectively.
During a discussion, ChatGPT mentioned it can process inputs in numerous languages, though the exact count varies based on dialects and regional variations. One demonstration highlighted its ability to correct French pronunciations, offering tips on inflection. In another instance, ChatGPT delivered an emotive story in Turkish, with some users noting that the accent didn't sound native, but it successfully completed the narrative, incorporating laughter and tears at appropriate moments.
The chatbot displayed competence with various regional US accents, including New York, Boston, and Wisconsin, as well as a stereotypical "valley girl" voice. Other demonstrations featured ChatGPT singing in different styles, delivering a bluesy rendition of "Happy Birthday" and humorously mimicking the singing voices of animals like frogs and cats.
A range of male and female voices were showcased, although the previously available "Sky" voice, reminiscent of Scarlett Johansson, was not included. OpenAI spokesperson Taya Christianson announced that the advanced voice mode will soon be accessible to all ChatGPT Plus subscribers, priced at $20 per month.