GPT-Live-1.1 has new voices and better accents.

Summary
This official OpenAI promotional video announces new voices and improved accents for ChatGPT Voice, demonstrating multiple voice personas across several languages. The video presents smartphones displayed on a theatre dressing table, cycling through roll calls in various languages before concluding with an end card introducing GPT-Live-1.1.

What is shown

• [0:00] Overhead view of two smartphones running ChatGPT Voice on a backstage dressing table lit with vanity bulbs.
• [0:03 - 0:28] Roll call of paired male and female ChatGPT voices conversing in multiple languages with corresponding animated voice orbs:
  • [0:03] French: “Okay, petit appel, tout le monde est prêt?” / “Je suis prête.”
  • [0:08] Spanish: “Aquí estamos.” / “Muy bien, ¿quién sigue?”
  • [0:12] German: “Schon immer bereit.” / “Ich höre euch laut und deutlich.”
  • [0:16] Korean: conversational check-in (“이쪽도 잘 들려요” / “다들 온 것 같네요”).
  • [0:20] Hindi: conversational check-in (“बढ़िया, चलें शुरू करते हैं” / “इतनी जल्दी नहीं”).
  • [0:24] Japanese: conversational check-in (“待って、誰か抜けてない？” / “え、なんか一人足りない気がする”).
• [0:30] A single phone pans into view with an English persona saying, “Sweetheart, I’ve been ready.”
• [0:34 - 0:40] Text overlay reads: “New voices are here. Find yours.”, “Try ChatGPT Voice”, and “Now powered by GPT-Live-1.1”.

Claims & numbers

• The video claims that ChatGPT Voice features new voices (“New voices are here. Find yours.”) [0:34].
• The video states ChatGPT Voice is “Now powered by GPT-Live-1.1” [0:39].

Notable quotes

• [0:00] “Alright everybody, it’s showtime!”
• [0:30] “Sweetheart, I’ve been ready.”
• [0:39] “Now powered by GPT-Live-1.1”

Assessment
This is an official, highly produced commercial launch teaser demonstrating vocal quality, natural delivery, and accent proficiency across multiple languages. It is a scripted marketing showcase rather than an interactive benchmark or unfiltered live user demo.

Described by gemini-3.8-flash on 2026-10-16 from the video’s audio and frames.