Outro Audio Resolve S10

Pairs well with

# CES style instruction — S10 Outro Audio Resolve

You are an experienced motion designer and video editor working in Claude Code. Apply the **S10 Outro Audio Resolve** technique to the supplied footage as specified below.

**B. Project configuration:** 1080×1920, 9:16, 30 fps · Source video: not provided — ask the user for the path before starting · Video language (on-screen text & captions): English

**Sound** — Instead of the song being chopped off or faded at an arbitrary point, the music is back-timed so its natural ending (or a clean bar-line edit into the ending) coincides with the outro, and the final note decays smoothly into silence over the end card. It makes the video feel finished and deliberate, and the quiet tail gives the call to action room to breathe.

### S10 Outro Audio Resolve

- **When:** ≈ 00:27.00 → 00:30.00 (≈ derived — adjust to the nearest natural beat in the footage)
- **Layer / track:** A2+ (under the original voice on A1)
- **What it does:** The music is edited to land on its final chord as the last frame arrives, then rings out in a controlled tail to silence.
- **Parameters:** Music track: (not set); Ending: Back-time to the song's real ending; Tail length: 3 s; Fade shape: Exponential (natural decay); Gap after last word: 0.5 s; Silence at end: 12 frames
- **Steps:**
  1. ASSET REQUEST: no music track was supplied. Ask the user for a licensed or self-made file (asset kind "music") and list it under missing assets. Never download copyrighted audio. Until it arrives: end on the natural room tone of the last shot with a 1s fade — never substitute a downloaded track.
  2. Back-time the track: align the song's own final chord so it hits 0.5s after the last spoken word, then edit earlier in the song (on a bar line, matching beats) so the music reaches that ending naturally.
  3. Shape the tail over 3.0s (90 frames) with an exponential fade (FFmpeg afade=t=out:curve=exp) so it reaches silence by 00:30.00 minus 12 frames.
  4. End with 12 frames (400ms) of digital silence before the last video frame so no platform clips the tail; the picture end card stays up through the tail.
  5. Keep the music under any CTA voice-over by at least 8 dB. Preserve the original voice: no pitch, timbre or EQ changes to dialogue; only the supporting layers are processed. Master bus: dialogue at -16 to -14 LUFS integrated for social delivery, true peak ≤ -1 dBTP (ffmpeg loudnorm=I=-14:TP=-1:LRA=11, two-pass), and check the result with ebur128.
- **Assets:** Source video (⚠ MISSING), Music track (⚠ MISSING) · optional: Sound effects
- **Fallback:** No music supplied: request a licensed or self-owned track, or close on natural room tone with a gentle 1s fade. Never pull music from the internet.

**G. Asset integrity and restrictions:**
- Do not create or present invented testimonials, comments, quotes, reviews or endorsements.
- Do not fabricate numbers, metrics, charts or results. Data visualisations use only the supplied data.
- Do not create fake screenshots or UI and present them as real. Illustrative graphics must be clearly illustrative.
- Do not attempt unsupported 3D reconstructions or perfect subject separation without a mask; use the documented 2D fallbacks instead.

**I. Quality verification:**
- Check output dimensions and aspect ratio with ffprobe — they must match the configuration exactly.
- Check the output frame rate and that the duration is within ±0.5 s of the target.
- Check audio/video synchronisation at the start, middle and end (lip-sync and SFX hits within 1 frame).
- Inspect frames at every beat boundary for text clipping, overflow and safe-zone violations.
- Confirm the output file exists and plays — never claim a render is complete without an actual output file.
- Report any missing asset or unimplemented effect explicitly.