2026-11-14 · 5 min read

How to add captions to ASMR videos with an API

A film strip of quiet studio scenes, the middle frame showing the captioned word ASMR in neon
This image was generated with AI.

ASMR is audio-first content, which makes captions sound pointless until you look at how people actually watch: in bed with the volume down, on a commute without headphones, or following a roleplay's story. Captions also help the sizable share of viewers who watch for relaxation but cannot catch whispered words at low volume. caption.sh generates them with an API call.

The workflow

import requests, time

API = "https://api.caption.sh/v1"
HEADERS = {"Authorization": "Bearer YOUR_API_KEY"}

job = requests.post(
    f"{API}/videos",
    headers=HEADERS,
    json={
        "source_url": "https://your-cdn.example/asmr-roleplay.mp4",
        "options": {
            "font": "inter",
            "position": "bottom",
            "max_words": 5,
            "max_lines": 2,
            "caps": False,
            "backdrop_opacity": 0.3,
        },
    },
).json()
job_id = job["job_id"]

for _ in range(120):
    job = requests.get(f"{API}/jobs/{job_id}", headers=HEADERS).json()
    if job["status"] in ("done", "error"):
        break
    time.sleep(5)

if job["status"] == "done":
    r = requests.get(f"{API}/jobs/{job_id}/result", headers=HEADERS,
                     allow_redirects=False)
    video_url = r.headers["Location"]

Jobs run queued, awaiting_upload, running, then done or error; poll until a terminal status. The result endpoint answers 302, so read the Location header with redirects disabled. Results are stored for 24 hours, then deleted.

Whispers and quiet audio

Whispered and soft-spoken voice is harder for any transcription than clear speech, and ASMR pushes that to the limit. Expect best results on soft-spoken roleplays and guided content with a clean mic, and plan a review pass on fully whispered tracks. Pure trigger videos, tapping, brushing, no talking, have nothing to caption and gain nothing from the API.

Styling for ASMR

  • Calm beats punchy here: caps off, max_words at 5, and a low backdrop_opacity keeps text unobtrusive over slow footage.
  • Bottom position keeps captions clear of face-focused framing.
  • For trigger assortments with spoken intros, caption the intro segment only.

Tradeoffs and cost

  • Burned-in captions cannot be toggled, resized, or translated. Keep platform captions enabled for viewers who need adjustable text.
  • Each job covers up to 10 minutes. Long ASMR sessions should be segmented, which also fits how channels cut compilations.
  • A 429 means the transcription throughput ceiling is reached. Wait and retry rather than submitting duplicates.
  • At $0.10 per minute prepaid, a 10-minute roleplay segment costs $1.00.

Back to all articles