Researchers at Arizona State University have discovered similarities in Scarlett Johansson’s voice and the now-discontinued GPT-4o voice "Sky" from OpenAI. According to NPR, the analysis commissioned by the outlet revealed that Johansson’s voice resembles Sky more than 98% of other actresses when compared using AI models developed for vocal analysis.

While Johansson’s voice was found to be similar to Sky, it was also noted that the voices of Anne Hathaway and Keri Russell had more resemblances to Sky than Johansson’s. Visar Berisha, the professor leading the analysis, emphasized that Johansson’s voice is similar but likely not identical. Berisha is also known for his work on OriginStory, a microphone that won an FTC voice cloning challenge.

The research team concluded that there are undeniable commonalities between Sky’s and Johansson’s voices, with their vocal tracts having identical lengths. However, some differences were observed, such as Sky’s voice being slightly higher-pitched and more expressive than Johansson’s, while Johansson’s voice was described as slightly more breathy than Sky’s.

OpenAI CEO Sam Altman and CTO Mira Murati have denied any intention for Sky to sound like Scarlett Johansson. Following the GPT-4o demo, where Altman cryptically posted "her," Johansson revealed that Altman had approached her to lend her voice to the model, a request she declined. Despite her refusal, Altman reportedly made another attempt to secure her voice just two days before the demo.

It is clear from the analysis that there are both similarities and differences between Sky’s and Johansson’s voices. While the voices may share some characteristics, each voice has its unique qualities that set it apart. The AI technology used for vocal analysis continues to evolve, shedding light on the fascinating intricacies of human speech and sound production.

As we delve deeper into the world of artificial intelligence and voice technology, the exploration of vocal resemblances and distinctions opens up new avenues for understanding human communication and expression. The study of voices like Scarlett Johansson’s and the AI model Sky provides valuable insights into the intricate workings of vocal patterns and their variations.

In conclusion, the comparison between Scarlett Johansson’s voice and the AI model Sky offers a compelling look into the complexities of vocal analysis and the nuances of human speech production. As technology advances, the ability to dissect and compare voices at a detailed level enhances our understanding of vocal similarities and differences, shedding light on the fascinating realm of voice recognition and artificial intelligence.