AssemblyAI-TTS-1, our first text-to-speech model. (Launch Video Storyboard & Reverse-Engineering Breakdown)

About AssemblyAI & This Product Release

AssemblyAI — Introducing AssemblyAI-TTS-1, our first text-to-speech model. Natural speech for voice agents, streaming in 90 ms, at $15 per million characters. On accuracy, naturalness, and cost together, it’s the best TTS model you can run at scale. Watch the full 90s AssemblyAI launch video with shot-by-shot breakdown and Remotion code.

Official X Account: @AssemblyAI | Original X Launch Post →

Explore AssemblyAI Company Profile & Video Archive (/company/assemblyai) | Visit Official AssemblyAI Website (AssemblyAI) →

Editorial Structure & Storyboard Timeline (90s Total Runtime · Fast (2.2s avg shot))

  • 01 · THE HOOK (00:00 – 00:13 · 14%): Prompt-to-Execution Split Reveal for AssemblyAI — Opens immediately on the high-contrast tension point before showing the brand logo, locking viewer retention in the first 13 seconds. (Camera: 1.6x macro push-in with 0ms dead air on frame 1)
  • 02 · HERO UI REVEAL (00:13 – 00:40 · 30%): Core AssemblyAI Surface & Interaction Loop — Demonstrates the primary action in a single continuous take (27s) with cursor smoothing and highlighted focal bounds. (Camera: Isometric-to-flat spring settle (cubic-bezier 0.22, 1, 0.36, 1))
  • 03 · VELOCITY MONTAGE (00:40 – 01:10 · 33%): Multi-State Feature Proof on Downbeat Cuts — Compresses 3–4 secondary workflows into rapid visual proof without narration drag. (Camera: Fast (2.2s avg shot) matched to 122 BPM transients)
  • 04 · PAYOFF & LOCKUP (01:10 – 01:30 · 23%): AssemblyAI Brand Lockup & Availability CTA — Holds final URL and launch status cleanly for 20s so viewers on X stop scrolling and click the link. (Camera: Clean center lockup with subtle radial luminance decay)

Release Context & Production Format

Published by AssemblyAI (@AssemblyAI) on 2026-10-28 in the ai category (90s total runtime).

30fps Remotion React Code Specification

import { AbsoluteFill, Sequence, useCurrentFrame, useVideoConfig } from "remotion";

// AssemblyAI — AssemblyAI-TTS-1, our first text-to-speech model. (90s @ 30fps)
export const AssemblyAIComposition = () => {
  const frame = useCurrentFrame();
  const { fps } = useVideoConfig();

  return (
    <AbsoluteFill style={{ backgroundColor: "#141413" }}>
      <Sequence from={0} durationInFrames={390} name="Act01_hook">
        {/* 00:00 – 00:13: Prompt-to-Execution Split Reveal for AssemblyAI */}
      </Sequence>
      <Sequence from={390} durationInFrames={810} name="Act02_reveal">
        {/* 00:13 – 00:40: Core AssemblyAI Surface & Interaction Loop */}
      </Sequence>
      <Sequence from={1200} durationInFrames={900} name="Act03_velocity">
        {/* 00:40 – 01:10: Multi-State Feature Proof on Downbeat Cuts */}
      </Sequence>
      <Sequence from={2100} durationInFrames={600} name="Act04_payoff">
        {/* 01:10 – 01:30: AssemblyAI Brand Lockup & Availability CTA */}
      </Sequence>
    </AbsoluteFill>
  );
};