Polaris
BlogGuide

ASMR-Style Ads with AI: A Practical Guide

How to generate ASMR-style UGC ads with AI using Polaris inside Claude, from script to finished, captioned video.

GuideJul 15, 2026·3 min read

To make an ASMR-style ad with AI, connect Polaris to Claude, then ask Claude in chat to generate a close-up, sound-focused ad from your product photo or script. Settings such as model and music are picked from tappable option cards, and the finished clip renders live in the conversation.

What makes ASMR ads work

ASMR-style ads lean on close, tactile visuals — texture, sound cues, slow deliberate motion — rather than a talking pitch. The appeal is sensory, which makes them a good fit for products with a satisfying physical quality: packaging, textures, pours, snaps, clicks.

Step-by-step: building the ad

  1. Connect the MCP once. Add polarisugc.com/api/mcp to Claude as a connector and authorize it.
  2. Attach a product photo. Close-up ASMR content works best from a photo that already shows texture or detail clearly.
  3. Describe the sensory beats in chat. Tell Claude the moments you want: a slow reveal, a texture close-up, a satisfying sound cue, a final product shot.
  4. Choose settings from the option cards. Pick from the available video models (Kling 3.0, Veo 3.1 Quality/Fast, Seedance 2.0, Omni) and decide whether to add Suno-generated music or keep it ambient.
  5. Let the live panel render it. The Polaris panel shows progress and the finished ad directly in the conversation, typically in 20-180 seconds.
  6. Add burned-in captions. Useful even on a sound-led format, since a large share of viewers watch muted first.
  7. Batch a few pacing variants. Slow vs. slightly faster cuts render in parallel, then download as a ZIP to compare.

Pro tips

  • Keep the script sparse — ASMR ads succeed on pacing and sound design cues more than dense copy.
  • Run Copy Judge on your script notes before rendering so the pacing instructions are clear and specific.
  • Try both a voiceover-led cut and a music-only cut of the same concept; they read very differently.
  • Use Recreate to quickly reproduce a cut that tested well with a new product photo.

Honest caveats

ASMR pacing is subjective — what feels satisfying to you may not test the same with a cold audience, so treat each cut as a hypothesis to test, not a finished answer. Polaris builds the ad around your product images and script; it doesn't draw from a stock-avatar library or clone a named real person's voice or likeness. If your product sits in a regulated category, keep any spoken or captioned claims compliant — no disease, cure, or guaranteed-outcome language, even in a soft-spoken ASMR tone.

Where the web app fits

The web app is a secondary surface for managing renders outside chat; the primary workflow is asking Claude directly once the MCP connector is set up, as covered in this walkthrough. Compare formats on the UGC ads page and see plans on pricing.

Frequently asked questions

What products suit ASMR-style AI ads best?
Products with a tactile or sensory quality — texture, pour, snap, click, unwrap — tend to work well since the format relies on close-up visuals and sound cues rather than a spoken pitch.
Do I need a separate sound designer for ASMR ads?
No. You can add Suno-generated music or keep the ad ambient, and burned-in captions are one click if you want the ad to read clearly muted.
How long does an ASMR-style ad take to render?
Renders typically finish in 20 to 180 seconds and show live progress in a panel inside the Claude conversation.
Can I test a music-led cut against a voiceover-led cut?
Yes, you can generate both versions in parallel and compare them once the ZIP download is ready.

Create your first winner now

One product link in. Winning ads out.

Keep reading

GuideMockup Ads vs. Real Renders: Which to Use WhenGuideGreenscreen-Style Ads with AI: Step-by-StepGuidePOV Ads with AI: How to Make Them Step by Step