Horror Trailer Sound Design & Captioning Tips

Pair cinematic horror sound design with practical SDH captioning to make trailers — like Legacy and Empire City — terrifying and accessible.

Make the chills count: Sound design for horror trailers that remains terrifying — and accessible

As a creator, you know that a trailer lives and dies by its sound. The right low rumble, a sudden high-frequency scrape, or the subtle tempo of a heartbeat can sell a scare in three seconds. But those sonic decisions often leave deaf and hard-of-hearing audiences out of the moment unless you caption audio cues correctly. This guide pairs modern horror trailer sound-design techniques (with examples inspired by 2026 releases like Legacy) and tense set-pieces like those in Empire City with structured captioning methods to make chills accessible — without diluting the craft.

Why this matters in 2026: industry and accessibility trends

Late 2025 and early 2026 saw two clear trends that affect how creators produce and ship trailers:

Wider adoption of spatial and object-based audio in streaming platforms (Dolby Atmos mixes are now common in high-profile trailers), which changes how sound cues move in the sound field.
Increased regulatory and platform pressure for high-quality SDH (Subtitles for the Deaf and Hard-of-Hearing). Major streamers and distributors pushed better captioning standards throughout 2025, and studios are asking for richer non-speech descriptions in deliverables.

That convergence means post-production teams must integrate captioning for audio cues into the mix, not bolt it on afterwards.

Two cinematic case studies: Legacy vs. Empire City (what their trailers teach us)

Referencing industry news from January 2026, David Slade’s Legacy (an atmospheric horror) and the hostage-thriller Empire City demonstrate two very different sound palettes — both useful when planning captioned audio cues.

Legacy (atmospheric horror)

What to learn: low-frequency drone, whisper placement, spatial ambiguity. Horror trailers like Legacy build dread through evolving textures rather than constant action. Your captioning needs to reflect subtlety — e.g., differentiate between a sustained drone and an approaching thud.

Empire City (action-hostage thriller)

What to learn: foreground SFX, directional cues, rapid dynamic changes. Trailers for tense thrillers escalate quickly: gunshots, doors slamming, footsteps converging. Captioning must handle fast, overlapping events with precision and clarity.

Core principles: marrying sound design with captioning

Follow these principles to keep the impact intact while ensuring accessibility:

Plan captions from pre-production — mark non-speech cues while scoring and sound-designing so captions are not afterthoughts.
Use consistent label conventions — pick a style (e.g., [SFX:], [MUSIC:], [VOICE-OFF], [WHISPER]) and use it through the project.
Prioritize timing over verbosity — captions should appear in sync with the cue; concise descriptions beat long sentences during fast edits.
Differentiate intensities — include markers like (low), (rising), (sudden) to cue emotional weight.
Respect spatial cues — for object-based mixes, indicate direction when it matters (LEFT, RIGHT, BEHIND).

Practical sound-design tips for horror trailers (with captioning outcomes)

Below are production and post tips that map directly to caption content. Each tip includes a short captioning template so editors can copy-paste.

1. Build dread with a layered drone

Sound design: Stack sub-bass, tonal grind, and an occasional harmonic scrape so the drone grows without overt changes. Automate subtle modulation to avoid repetitive loops.

Captioning: Time the caption to when the drone becomes perceptible and show intensity changes.

[00:00:02.400 --> 00:00:06.200]
[SFX: low drone (rising)]

2. Use whispered voice as texture (not always legible)

Sound design: Place whispered lines subtly in the mix with reverb tails. Use varied panning to make the whisper feel like it moves around the camera.

Captioning: If whispered content is intelligible, caption the words at lower opacity labels; if not, describe:

[00:00:10.100 --> 00:00:11.800]
[WHISPER: indistinct]

3. Design jump-scare sequencing with pre- and post-cues

Sound design: A jump-scare works best when your audience expects it—use a pre-cue (tension build) and then an impact spike. Keep the impact short but high in frequency energy.

Captioning: Break the caption into two: the build and the impact. That keeps the visual reader aligned to the audio arc.

[00:00:21.000 --> 00:00:22.600]
[SFX: tension build (metallic scrape, rising)]

[00:00:22.600 --> 00:00:22.900]
[SFX: impact (loud bang)]

4. Make breathing a character

Sound design: Isolated breaths can be terrifying when mixed close. Use slight compression and a touch of proximity EQ to simulate intimacy.

Captioning: Vary the description by pace.

[00:00:14.000 --> 00:00:15.200]
[SFX: breath (shallow, rapid)]

5. Spatialize footsteps and offscreen sounds

Sound design: Use panning automation and subtle delay to create directionality. In Atmos/object mixes, assign footsteps as objects to move across the field.

Captioning: Capture direction and proximity.

[00:00:27.000 --> 00:00:28.400]
[SFX: footsteps (approaching from LEFT, heavy)]

6. Score minimally — let SFX do the heavy lifting

Sound design: In trailers like Legacy, sparse scoring allows SFX to punctuate moments of dread. Use music as texture, not as loud cover for dialogue or important cues.

Captioning: When music competes with audio cues, call it out.

[00:00:05.000 --> 00:00:09.000]
[MUSIC: low string drone (ominous)]

Captioning mechanics: formats, exports, and deliverables

Deliverables differ by platform. Here's what to prepare for 2026 workflows:

SRT — Good for web and social, but lacks robust metadata for SDH styling. Use for fast turnaround and social clips, but supplement for accessibility-critical deliverables.
SCC / CEA-708 — Broadcast closed caption format that supports positioning and style; required for traditional TV deliverables.
TTML/DFXP — Preferred for rich styling, used by many streaming platforms. You can embed extended cue descriptions and styling sheets.
WebVTT — Modern web standard that supports cue settings (line positioning). Good for HTML5 players when paired with ARIA practices.

Best practice: always export a rich TTML or DFXP caption track containing the full set of non-speech descriptions, then produce SRT/WebVTT variants for platforms that require them.

SDH style guide: concise rules for teams

Standardize language across teams to reduce edits and ensure compliance. Use this mini style guide in your post-production SOP.

Label categories: [SFX], [MUSIC], [VOICE], [WHISPER], [OFFSCREEN], [PHONE], [AUDIO LOG]
Direction: Use all-caps for direction words (LEFT, RIGHT, BEHIND, OFFSCREEN)
Intensity: Use parenthetical modifiers — (soft), (urgent), (distant), (loud)
Length: Keep lines to two short phrases for fast pacing. Avoid line wrapping beyond 32–40 characters where possible.
Overlap: If multiple SFX overlap, stack them on separate caption frames or use slashes to indicate simultaneous events: [SFX: whisper/heartbeat]
Non-literal sounds: For ambiguous cues, prefer descriptive labels: [SFX: mechanical groan] vs. conjecture like [SFX: ghost]

Caption templates for common horror trailer cues

Copy these templates into your captioning tool or transcript editor. Adjust timing to match your EDL/Timeline.

Template: Slow-burn opening

[00:00:00.000 --> 00:00:05.000]
[MUSIC: low drone (ominous)]

[00:00:02.000 --> 00:00:03.000]
[SFX: distant child laughing (faint)]

[00:00:04.600 --> 00:00:05.200]
[VOICE-OFF: "It remembers."]

Template: Rapid jump-scare sequence

[00:00:12.000 --> 00:00:12.600]
[SFX: scrape (sharp)]

[00:00:12.600 --> 00:00:13.000]
[SFX: impact (loud bang)]

[00:00:13.000 --> 00:00:15.000]
[MUSIC: hit (stinger), then silence]

Template: Spatialized footsteps and doors

[00:00:20.400 --> 00:00:21.800]
[SFX: footsteps (approaching from RIGHT)]

[00:00:22.000 --> 00:00:22.600]
[SFX: door creak (slow, from LEFT)]

Workflow checklist: integrate captioning into post-production

Use this checklist to make captioning non-disruptive to your mix and edit cadence.

Pre-mix session: Create a cue sheet of anticipated non-speech events while sound design begins.
Temp captions in editorial: Have editors insert placeholder captions (using your style guide) during picture lock.
Sound mix pass: Finalize cue timing and intensities in the mix; update caption timecodes accordingly.
Caption engineering: Export final captions to TTML/DFXP and test them in the deliverable player (streaming, social, broadcast).
Quality assurance: Run a pass with deaf or hard-of-hearing reviewers where possible; prioritize readability and emotional accuracy.
Deliver: Provide master video with embedded captions and separate caption files in required formats.

Tools and AI-assisted options (2026 landscape)

By 2026, AI has matured in audio transcription and description generation. But human oversight remains essential for creative nuance.

Automated transcription & captioning: Tools like Descript, Trint, and cloud ASR (Google, AWS, Azure) accelerate dialogue captioning. They can also suggest non-speech labels, but expect noise in ambiguous cues.
Audio repair and separation: iZotope RX and AI-driven source separation speed isolation of whispers and breaths for clearer caption placement.
Spatial audio authoring: Dolby Atmos and immersive DAWs let you export object metadata; use that to inform caption directions.
Caption editors: Use specialized caption tools (EZTitles, SubtitleNEXT, CaptionHub) for precise timing and format exports like SCC/TTML.

Tip: Use AI to get a first draft of non-speech descriptions, then have a sound editor or accessibility specialist review and refine for emotional intent.

Testing and user feedback: key to emotional accuracy

Accessibility is not only legal compliance — it's creative integrity. Test your captions with people who are deaf or hard-of-hearing and ask specific questions:

Do the audio cue captions convey the emotional arc?
Are overlapping events readable and intelligible?
Do direction and intensity descriptors match perceived sound position?

“Captions should carry the trailer’s emotional spine, not just its literal audio events.”

Common pitfalls and how to avoid them

Too much text: Avoid verbose descriptions during fast edits. Use concise action words and intensity markers.
Mismatched timing: Sync captions to the first perceptible audio cue, not the edit cut, for natural reading alignment.
Over-interpretation: Don’t label ambiguous sounds as supernatural unless context supports it. Describe the sound instead.
Ignoring spatial mixes: If you’re shipping Atmos, include direction tags; viewers of stereo will still benefit from the cue text.

Final checklist before delivery

All non-speech cues labeled with one of your standard tags.
Intensity and direction markers included where they alter perception.
Caption files exported in TTML/DFXP and SRT/WebVTT variants.
QC pass with at least one deaf/hard-of-hearing reviewer or accessibility consultant.
Player test on target platforms (mobile, desktop, TV) to confirm positioning and legibility.

Actionable takeaways

Start caption planning at pre-production; make it part of the cue sheet.
Use concise, consistent labels and intensity descriptors to preserve emotional impact.
Export rich caption formats (TTML) for distribution and lightweight formats (SRT) for social.
Leverage AI tools for drafts, but always run a human review focused on emotion and clarity.
Test with deaf and hard-of-hearing users to validate your creative choices.

Closing: make every scare inclusive

Trailers can be the most public-facing piece of a film’s marketing. In 2026, with rising expectations for quality SDH and immersive audio formats, creators must design sound and captions in parallel. Whether you’re crafting the slow dread of a David Slade-esque film like Legacy or the high-intensity spikes of an Empire City-style sequence, the right captioning approach preserves the emotional intent for deaf and hard-of-hearing audiences without compromising craft.

Start by adopting a simple captioning style guide, integrate caption creation into your post workflow, and treat captioning as a creative responsibility. The result: trailers that land harder and reach further.

Call to action

Ready to ship trailers that frighten and include? Download our free SDH caption templates and a one-page style guide optimized for horror trailers at descript.live/resources. Try embedding these templates into your next edit session and schedule a QC review with a deaf or hard-of-hearing consultant before final delivery — you’ll protect the scare and expand your audience.