Startup Launch · Launch Video Breakdown: Hook, Pacing & Motion Design
New voices. Better accents. And less "checking..." GPT-Live-1.1 brings these updates to Voice in ChatGPT, Codex, and dots, with new voices in French, German, Hindi, Mexican Spanish, Korean,
Scene-by-scene timeline & spoken transcript
The Hook
Curtain Call Kickoff
“Alright everybody, it's showtime! Okay! Quick roll call. Everybody ready? Je suis prête.”
- On screen
- Alright everybody, it's showtime! Okay! Quick roll call. Everybody ready? I'm ready.
- Camera
- Top-down overhead static framing, beginning a lateral tracking pan.
- Motion
- Pulsing audio orb animations on mobile screen interfaces.
Feature Teaser
Spanish Check-In
“Aquí estamos. Muy bien. ¿Quién sigue?”
- On screen
- We're here. Great. Who's next?
- Camera
- Smooth horizontal rightward pan across marble tabletop.
- Motion
- Teal fluid waveform activation synchronized with audio.
Feature Teaser
German Voice Pair
“Schon immer bereit. Ich höre euch laut und deutlich.”
- On screen
- Born ready. Coming through loud and clear.
- Camera
- Continuous steady lateral dolly right.
- Motion
- Monochrome charcoal visualizer reactive pulsing.
Feature Teaser
Korean Voice Pair
“이쪽도 잘 들려요. 다들 온 것 같네요.”
- On screen
- Everything sounds good over here. Looks like the whole team's here.
- Camera
- Continuous lateral pan right.
- Motion
- Warm golden visualizer sphere morphing.
Feature Teaser
Hindi Voice Pair
“बढ़िया चलो शुरू करते हैं. इतनी जल्दी नहीं.”
- On screen
- Perfect. Let's do this. Not so fast.
- Camera
- Continuous tracking pan past vanity lights and red curtains.
- Motion
- Coral pink waveform bloom.
Problem Agitation
Japanese Missing Voice
“待って、誰か抜けてない？ え、なんか一人足りない気がする.”
- On screen
- Wait. Are we missing someone? I think we're one short.
- Camera
- Lateral tracking decelerates slightly.
- Motion
- Purple visualizer sphere ripple.
Product Reveal
Solo Voice Reveal
“Sweetheart, I've been ready.”
- On screen
- Sweetheart, I've been ready.
- Camera
- Pan settles and centers on the final lone phone.
- Motion
- Deep blue audio orb expanding in time with dramatic line delivery.
Call to Action
Title & Engine Branding
“(No spoken dialogue — orchestral musical flourish concludes)”
- On screen
- New voices are here. Find yours. Try ChatGPT Voice Now powered by GPT-Live-1 Now powered by GPT-Live-1.1
- Camera
- Static graphic cards.
- Motion
- Staggered text fade-ins followed by kinetic model version flip.
Motion design primitives
Synthesizable in Remotion · key components of the lateral-ensemble-pan template.
LateralTableTracking
Seamless horizontal linear pan across tabletop props and paired devices.
+ Code construction recipe
Translate container horizontally via interpolate(frame, [0, totalFrames], [0, -maxOffset], { extrapolateRight: 'clamp' }).ReactiveVoiceOrb
Radial gradient circle expanding and distorting with speech audio energy.
+ Code construction recipe
Use spring-driven scale combined with Perlin noise SVG displacement based on frame audio timestamps.
KineticSubtitleDisplay
Centered bottom subtitles appearing smoothly per dialogue cue.
+ Code construction recipe
Conditional rendering mapped to timestamp array with opacity fade using spring physics.
Style DNA & aesthetic system
Visual aesthetic
Typography
Motion
Special effects
Pacing & rhythm
Conceptual story rhythm
- 1
Roll Call Launch
Purpose. Establish premise and lively multi-agent banter.
Execution. Top-down overhead view of initial phone pair.
- 2
Multilingual Sweep
Purpose. Demonstrate global voice diversity.
Execution. Continuous pan across distinctly colored visualizer pairs.
- 3
Comedic Cliffhanger & Resolution
Purpose. Build anticipation for the final distinct voice.
Execution. Tracking reveals the final isolated device delivering the punchline.
- 4
Call to Action
Purpose. Direct viewers to experience new capabilities.
Execution. Clean white card typography and engine disclosure.
Remix in Animatiq Studio
Master remix prompt · agent-ready
Sample remix
Objective
Announcing regional accents for a multilingual customer service AI.How the recipe adapts
Replace dressing table with executive office desks, panning across tablets highlighting localized enterprise voice accents.Remix blueprint · lateral-ensemble-pan
Scene durations (4 scenes, 40s)
Key Remotion components
Related AI & ML Product Launches
Explore all AI & ML launches →
Introducing Griffin, the first AI to pass the video Turing test
The most emotive and controllable TTS model
We just released QVAC TranscribePsy, a speech-to-text model small enough to run inside a phone app. It covers 25 European languages on a laptop and 8 on a phone, and on an iPhone 17 Pro Max's Ne