OpenAI has updated ChatGPT’s Advanced Voice Mode feature, promising a more natural conversation experience. The aim is to make the AI-powered assistant more pleasant to talk to and less prone to interrupting you mid-sentence.
In a video posted on the OpenAI YouTube channel on Monday, researcher Manuka Stratta showed off the improvements. One of the most common annoyances with voice assistants is that they tend to interrupt you when you pause to think. That’s now been fixed here.
Free ChatGPT users will now have access to an updated version of the voice mode where they can speak freely and pause without being interrupted. Paid ChatGPT users will also get a voice assistant with better personalization and fewer interruptions, TechCrunch reports.
According to OpenAI, the improved chatbot is now “more direct, engaging, concise, specific, and creative in its responses.”
The update comes at a time when competition is heating up in the AI voice space. Sesame, founded by Brendan Iribe (formerly of Oculus), has recently gotten a lot of attention for its naturally speaking assistants Maya and Miles. It was so good that it actually freaked out PCWorld’s Mark Hachman when he gave it a try. Amazon is also working on a new version of Alexa that’s based on a large language model.
In other words, AI chatbots and assistants are not only getting better at hearing, but also learning to listen and response more naturally.