AI Video Editing Flows

Reel teardown #21: a still-panel horror reel that uses speech bubbles and long crossfades instead of motion

We took apart one Instagram reel by @toonstorytv_darkx, a still-panel horror reel that uses speech bubbles and long crossfades instead of motion. It had 687,350 plays on 5 Nov 2026, 12.3x its page's median. Length, cuts, pace and loudness were measured by script, and the frames were read one by one.

The reel in numbers

The reel we took apart. Posted Oct 31, 2026 Open on Instagram (opens in a new tab)

Reel by @toonstorytv_darkx on Instagram. Watch the original reel. We link to it and do not copy it.

Posted on 31 Oct 2026. Plays and followers as shown on Instagram on 5 Nov 2026.

MeasureValueWhat it means
Length36.1 sLength of the file we measured.
Plays when measured on 5 Nov 2026687,350Plays as shown on Instagram that day. They include replays.
The page's median plays over its last ~10 reels55,670Plays of a typical recent reel on this page.
This reel against that median12.3xThis reel's plays divided by the page median.
Followers110,527Followers of the page on 5 Nov 2026.
Hard cuts per minute0.0Hard cuts counted by script. Fast dissolves are not counted.
Median shot length36.06 sMedian time between hard cuts.
First hard cut atNo hard cut foundTime of the first hard cut the script found.
Hard cuts in the first 3 s0Hard cuts found before the 3 s mark.
Big visual changes per minute5.0Hard cuts plus large camera travel or scene changes.
Integrated loudness-14.1 LUFSIntegrated loudness of the audio track, in LUFS.
SpeechNo speech detectedWords per second and a language guess from a speech-to-text pass.

The first three seconds

What we saw frame by frame: Static comic frame: a woman leaning on a balcony looks at a little girl standing at the gate at night; a Hindi speech bubble is on screen from frame 0; a dissolve at 3 s.

By script, there is no hard cut in this reel and 0 big visual changes in the first 3 s.

The reel opens on a static comic frame with a speech bubble already on screen. The first change is a dissolve at 3 s.

Captions and on-screen text

What we saw frame by frame: Hindi speech bubble beside character, white oval, black bold Devanagari, medium-large, full sentence; later bubbles stack with dissolve; logo badge bottom-right.

Speech bubbles carry the dialogue: a white oval with bold Devanagari text, a full sentence at a time. Later bubbles stack with a dissolve, and a small logo badge sits in the bottom corner.

Shots and pace

What we saw frame by frame: Still panels with long crossfades; one cut counted over 36 smeasured.

By script: 1 shot, 0.0 hard cuts a minute, a median shot of 36.06 s, and 5.0 big visual changes a minute.

The panels are stills joined by long crossfades. The script counts no hard cuts because a crossfade is not a hard cut, and the big-visual-change counter registers some of the changes. The overlapping dissolves leave ghosting in the notes.

Sound

What we saw frame by frame: Licensed spooky score 'Halloween: Whispers Of The Haunted', little to no speech.

Measured loudness: -14.1 LUFS. Speech: No speech detected.

A spooky score plays under the stills, with little or no speech.

Structure and ending

  • Structure: Woman spots girl then goes down to ask then girl's reply line then mystery beat (girl vanishes).
  • Ending: Ends on a ghostly double exposure; series cliffhanger.

A woman spots a girl at the gate, goes down to ask, the girl replies, and the girl vanishes. It ends on a ghostly double exposure as a series cliffhanger.

Effects we have for this

Speech bubbles and crossfades between stills are not close matches to anything on our effect list. Browse all our effects for the nearest options.

Method and limits

  • Length, size and frame rate: read from the file.
  • Hard cuts, shot length and first cut: a script that finds hard cuts between frames. Fast dissolves are missed.
  • Big visual changes: hard cuts plus large camera travel or scene changes.
  • Loudness: integrated loudness of the audio track in LUFS. It is blank when the file had no audio track.
  • Speech: words per second and a language guess from a speech-to-text pass.
  • Plays, followers and the page median: collected by Apify on 5 Nov 2026.
  • What we saw frame by frame: a model read a contact sheet of the reel. Treat these as observations, not measurements.
  • Hard-cut counts miss dissolves, so a reel built on dissolves can show close to zero hard cuts.
  • The big visual change count also counts camera travel and scene changes, so it is not the same as the cut count. We show both.
  • Some full-HD files had no audio track, so loudness and speech are blank for those reels.
  • Plays include replays. Shares and follower conversion cannot be seen from outside.
  • Ten reels from one page are not ten independent tests.
  • Hashtag discovery favours pages that already had a hit, so breakout pages are over-represented.
  • Plays per follower favours small, newer pages; it measures reach health, not size.
  • Formats measured on only one to three pages are hypotheses.
  • No number here is a promise about what a new reel will do.

Plays and followers as shown on Instagram on 5 Nov 2026. We link to the reel and do not copy or embed it.

Page checked on Nov 4, 2026.