Models · 15 October 2026 · 1 min read
GPT-Live-1.1: new voices for ChatGPT Voice
Rolling out from 15 October 2026: more natural accents and new voices in six languages, inside ChatGPT rather than the API.
By The Crewnema team
On 15 October 2026 OpenAI began rolling out two voice models in ChatGPT, GPT-Live-1.1 and a variant called GPT-Live-1.1 Quip. This is what OpenAI’s release notes said.
What OpenAI announced
- More natural-sounding accents in ChatGPT Voice.
- New voices for six languages.
- A rollout that depends on plan, region and account.
- No change for developers: the voice model in the API stays GPT-Live-1.
Our view
This one is for people talking to ChatGPT, not for people making adverts. A business cannot send a script to GPT-Live-1.1 and get a voiceover file back, because it is not in the API. Its interest for us is as a signal: accents are the part of synthetic voice that still gives it away, and every large maker is now working on them. Expect the same improvement to reach the voices you can actually use for a voiceover.
Side by side
In Crewnema
Voiceovers in Crewnema are made by the voice models on our roster, not by ChatGPT, and you can also record the line yourself and attach it to a shot.
Sources
Make your first video
Crewnema is in private beta and lets businesses in a few at a time. Request access and we will write when yours is open.