Summary
OpenAI has unveiled a new voice cloning technology called Voice Engine that can replicate a person’s voice with just a small sample of original audio. The technology has already been used to help a young patient with a brain tumor regain her ability to speak. OpenAI is being cautious about releasing the technology due to concerns about deepfakes and is working on ensuring responsible deployment. The company is also working on other projects such as GPT-5 and the generative video tool Sora.
Key Points
1. OpenAI has developed a new voice cloning technology called “Voice Engine” that can replicate a person’s voice, intonation, and speech patterns based on a relatively small sample of original audio.
2. The Voice Engine technology is able to create emotive and realistic voices with just a single 15-second sample, which is a significant advancement compared to other AI voice cloning tools that require longer audio samples for best results.
3. OpenAI is being cautious about the broader release of Voice Engine due to concerns about the potential misuse of synthetic voices, such as deepfakes. The company is implementing restrictions on the technology, including a list of prominent people it will not emulate, and is seeking explicit and informed consent from the original speaker before using their voice.