Launch Video Library · AI · Feature · 2026

GPT-Live-1.1 Voice Update

OpenAI introduces GPT-Live-1.1, bringing improved accents, reduced latency, and new voices across multiple languages to ChatGPT and Codex.

What OpenAI shipped

OpenAI announced GPT-Live-1.1, an update bringing new voices, better accents, and reduced latency to the voice interfaces of ChatGPT, Codex, and dots. The release expands language support to include French, German, Hindi, Mexican Spanish, Korean, and Japanese.

For a feature focused entirely on audio and latency improvements, the launch film must visualize the invisible. By relying on motion design to represent voice quality and speed, an announcement can communicate technical upgrades without requiring a lengthy technical explanation.

How the motion works

When introducing audio-centric features like new accents or reduced latency, a motion designer might rely heavily on kinetic typography to visualize the spoken word. Syncing text animations directly to the cadence of the new voices allows the viewer to 'see' the improved naturalness of the speech. A typewriter effect can be used to emphasize the speed of transcription, matching the exact pacing of the spoken audio to highlight the reduced 'checking...' delay.

To showcase multiple languages—like French, German, Hindi, and Japanese—cutting on the beat of a unified audio track can stitch disparate voice samples into a cohesive narrative. A designer could try using hard cuts between different typographic treatments or waveforms, each styled to subtly reflect the language being spoken, ensuring the transitions feel deliberate and rhythmically anchored.

Visualizing latency improvements requires careful attention to hold times and easing. To demonstrate the absence of delays, a designer might contrast a slow, ease-in animation representing the old 'checking' state with a snappy, linear overshoot animation for the new GPT-Live-1.1 response time. This stark contrast in motion curves instantly communicates the performance upgrade without needing a single chart or graph.

What to steal from it

  • Use kinetic typography to visualize audio pacing and natural cadence.
  • Sync visual cuts to the beat of spoken dialogue when showcasing multiple languages.
  • Contrast slow easing with snappy overshoot animations to demonstrate latency improvements.
  • Rely on the typewriter effect to emphasize real-time transcription speed.

Want a video like this for your product?

Impractical cuts one from a single prompt — the same motion craft, in about twenty minutes.