OpenAI’s new Voice Engine tool could have far-reaching consequences, as it enables the recreation of a person’s voice with just 15 seconds of recorded audio. Introduced in 2022, Voice Engine utilizes a short clip to learn the unique characteristics of a speaker’s voice and speech patterns. Users can then input text, and the tool will generate realistic-sounding voices with appropriate emotions.
While Voice Engine presents numerous potential benefits, such as aiding in presentations and improving communication, there are concerns about its misuse for deceptive purposes. Recognizing this risk, OpenAI has expressed a cautious approach and emphasized the importance of responsible deployment to prevent malicious activities.
To address these issues, OpenAI has engaged with various entities, including governments, media, entertainment, and educational institutions, to gather feedback on Voice Engine. Test users are prohibited from impersonating others and must disclose that the voice is AI-generated. Additionally, OpenAI has implemented watermarking to indicate the synthetic nature of the voice.
Moving forward, OpenAI intends to evaluate the feedback received to determine the future availability of Voice Engine. Regardless of the outcome, the company stresses the need for public awareness and understanding of the evolving technology landscape. Whether or not Voice Engine is widely released, OpenAI emphasizes the importance of transparency and responsible use in the development and deployment of synthetic voice technology.
