Launchscene
PricingSign in
Explore

Rissa Cao - Introducing @FishAudio Live 3: a full-duplex voice model that listen

Fish Audio · Nov 2026 · 2:12 · AI-annotated

Length
2:12
Scenes
24 · 5.5 s avg
Product on screen
0:00
Call to action
1:52 · 85%
Voiceover
Yes
00:00:00 / 02:12:01

Fish Audio’s 2:12 launch film runs 24 scenes, about 5.5 s each. The product is on screen from the first frame. The call to action lands at 1:52 (85% in), over a voiceover. The video opens with a one line launch, then runs a single real time demo of ordering at a McDonald's drive through, layered with frosted chat cards that type the conversation word by word. It then explains the full duplex architecture with simple frosted diagrams, closes with a pricing offer and a comment based call to action, and signs off on a peach gradient with the animated waveform mark. The signature moves are the live word by word typing in the translucent transcript cards, the lower第三rd

Footage neededEstimated needs: 2 capture groups — your screens or recordings · 1 official logo — reused across the film · 17 filmed passages — shoot, supply or replace
Plan my film from this onePrintable storyboard

Is this your launch? to your site or press kit.

More like this

Rissa Cao - Introducing @FishAudio Live 3: a full-duplex voice model that listen, scene by scene

· 24 scenes as text

  1. 01 · Product intro · 0:00–0:04

    A young man in a black t-shirt sits in a car's back seat, turning to face the camera and speak. A waveform icon with the title Live 3 and the line Full Duplex Voice Model fades in over the left seat, then out, as bottom subtitles read the launch line.

    On-screen text: “Live 3”, “Shija Liao”, “Co-Founder & Chief Scientist”, “Full Duplex Voice Model”, “Today we are launching Live 3”, “our full-duplex voice model”

    Techniques: Kinetic type

    Camera: Push in

  2. 02 · Demo · 0:04–0:06

    Three people sit in a car, two up front and one in the back, conversing. White subtitles caption their dialogue along the bottom edge, and a name lower-third labels the woman on the right. The camera stays locked off inside the cabin, a McDonald's visible through the rear window.

    On-screen text: “Rissa Cao”, “Co-Founder & CEO”, “hey I'm hungry”, “can we order some food?”, “sure”

    Camera: Static

  3. 03 · Demo · 0:06–0:09

    A hand holds an iPhone inside a car, its screen showing a call to Fish McDonald with a counting timer. A dark speech bubble on the right types out the assistant's reply beneath a pulsing waveform icon.

    On-screen text: “let's do it”, “Thanks for calling McDonald's, what can I get started for your pickup order?”

    Camera: Static

  4. 04 · Demo · 0:09–0:15

    A young man in a car holds a phone and speaks, while a translucent transcript card on the right types out his words word by word in a chat bubble. A second bubble labeled Fish Agent, reading Um-hm, fades in below. The steering wheel blurs the foreground.

    Camera: Static

  5. 05 · Demo · 0:15–0:24

    A young man in a black t-shirt sits in a car holding a blue phone, a second person in a white shirt behind him, while a translucent panel on the right types a live transcript word by word in a dark bubble over a blurred passenger.

    Techniques: Kinetic type

    Camera: Static

  6. 06 · Demo · 0:24–0:25

    A hand holds a phone in a car, its screen showing a voice memo titled Fish McDonald with a running timer. A cream chat bubble labeled Jiahua Liu slides in from the right and fills with the line Can I get another one, the same one?

    On-screen text: “Jiahua Liu”, “Can I get another one, the same one?”

    Techniques: Kinetic type

    Camera: Static

  7. 07 · Demo · 0:25–0:28

    A woman in a white sweater and a man in the back seat sit in a car while a blurred hand holds a blue phone in the foreground. A translucent Live Transcript panel on the left fills with chat bubbles as the order is spoken.

    Camera: Static

  8. 08 · Demo · 0:28–0:39

    A back-seat car shot with a woman in a lavender sweater talking into a phone, while a translucent live-transcript panel on the left types her speech into chat bubbles in real time, scrolling with the agent's replies.

    Techniques: Split screen, Kinetic type

    Camera: Static

  9. 09 · Demo · 0:39–1:06

    In a car at a McDonald's drive-through, a woman and a man hold phones while a translucent Live Transcript chat card overlays the frame. Chat lines type out word by word, focus shifts from the woman to the man, and an order confirmation receipt card rises into view.

    Techniques: Kinetic type

    Camera: Static

  10. 10 · Demo · 1:06–1:16

    Two men sit in a car at a McDonald's drive-through, the front passenger holding a phone. On the right, a frosted live-transcript panel scrolls: a reply bubble slides in, then the agent's message types out the order number and callback number, above a static order-confirmation card.

    Techniques: Kinetic type

    Camera: Static

  11. 11 · Other · 1:16–1:21

    A young man in a white T-shirt sits in a car's back seat holding a laptop, talking to the camera. Three frosted rounded boxes labeled ASR, LLM, and TTS fade in on the left, linked by white nodes and an orange dashed line that travels the path.

    On-screen text: “ASR”, “Automatic Speech Recognition”, “LLM”, “Large Language Model”, “TTS”, “Text-to-Speech”

    Camera: Push in

  12. 12 · Other · 1:21–1:33

    A man in a white tee talks to camera from a car back seat while a frosted Live 3 card sits left. Pill labels Speak, Listen, Interject slide in from the right and link to the card with lines; the shot widens to add Fish Agent Platform and API.

    On-screen text: “Live 3”, “Full Duplex Voice Model”, “Speak”, “Listen”, “Interject”, “Fish Agent Platform”, “API”

    Camera: Pull out

  13. 13 · Pricing · 1:33–1:52

    A woman in the driver's seat of a car parked by a McDonald's talks to the camera through the open door. White overlays fade in and out on the left: 50% off, Enterprise, a Custom Voice Agent card, FREE; her name and title sit lower right.

    On-screen text: “Live 3”, “Rissa Cao”, “Co-Founder & CEO”, “50% OFF”, “UNTIL END OF YEAR”, “Full Duplex Voice Model”, “ENTERPRISE”, “Book a demo”, “Custom Voice Agent”, “FREE”

    Techniques: Kinetic type

    Camera: Pull out

  14. 14 · CTA · 1:52–1:54

    A light-lavender chat screen holds a rounded photo of a woman in a car seat, looking back at the camera. A timestamp and small avatar sit beside a sender name that types out to 'Customer support agent'. Toolbar icons and a black Send button line the bottom.

    Camera: Pull out

  15. 15 · CTA · 1:54–1:54

    A social post composer on a light gray background frames a video of a woman seated in a car. Below the clip, a timestamp and avatar sit above a text field where the words “AI receptionist” type in, with a toolbar and Send button beneath.

    Camera: Static

  16. 16 · CTA · 1:54–1:55

    A messaging card displays a video thumbnail of a woman talking inside a car. The thumbnail scales up while a pill-shaped bubble with three dots types out a message across the bottom edge of the video.

    Techniques: UI zoom

    Camera: Push in

  17. 17 · CTA · 1:55–1:56

    A woman in a lavender sweater sits in the driver's seat of a car, smiling and speaking toward the camera. A dark rounded caption pill with a link icon holds the line at the bottom of the locked-off frame. A second person is partly visible in the back seat.

    On-screen text: “We'll send you the link”

    Camera: Static

  18. 18 · CTA · 1:56–2:01

    A soft peach gradient holds a faint vertical audio waveform in the center. The black title Live 3 sits above two lines that wipe in from left to right, one after the other, then hold static.

    On-screen text: “Live 3”, “Full Duplex Voice Model”, “Available now on fish.audio”

    Techniques: Kinetic type, Masking

    Camera: Static

  19. 19 · CTA · 2:01–2:03

    A young man in a black tee and dark pants walks across a sunlit parking lot holding a paper bag, the camera tracking him past parked cars and a McDonald's. A static text overlay stays locked to the frame throughout the shot.

    On-screen text: “Live 3”, “Full Duplex Voice Model”, “Available now on fish.audio”

    Camera: Pan

  20. 20 · Other · 2:03–2:04

    A static interior shot of a parked car shows three young people with fast-food takeout bags. The man left leans over a brown paper bag, the woman right tears open a white bag with orange and red print, and a third holds a bag in the back seat.

    Camera: Static

  21. 21 · Other · 2:04–2:06

    Inside a car, a boy in the front left seat holds a burger in a white box while a girl beside him bites a fry from a white paper bag. A third person in the back seat holds a brown bag. The camera stays locked off, framed through the windshield at a building outside.

    Camera: Static

  22. 22 · Other · 2:06–2:07

    Inside a car with a glass roof, a young man in a black shirt holds a burger box on the left, a woman in a lavender shirt eats a chicken piece on the right, and a third person sits in the back. The locked-off frame captures all three.

    Camera: Static

  23. 23 · Outro · 2:07–2:10

    The Fish Audio waveform mark sits centered on a soft peach gradient, its bars animating as the wordmark fades in from faint to solid black. The full-frame gradient then collapses into a horizontal band bordered by black bars.

    Camera: Static

  24. 24 · Outro · 2:10–2:12

    A black frame holds for most of the scene, then the white brand name Fish Audio appears centered and remains static for the final beat.

    On-screen text: “Fish Audio”

    Camera: Static