Short Drama Subtitle Removal

Why this matters

Vertical short dramas are localized at scale: 20 languages, one episode master. Burned-in Chinese subtitles block dubbing and force re-renders.

The Subtitle Eraser API detects subtitle regions per scene, inpaints the underlying video, and exports a clean master for the localization pipeline.

How it works

Point the API at the episode, let region detection find the text zones, and get back a subtitle-free master — plus a sidecar of detected regions for QC.

Submit a job with an input URL (S3, GCS or HTTPS), poll the job URL, and download the rendered output. No GPU, no queues, no ffmpeg builds to babysit.

Code

import requests, time

API = "https://api.mlslabs.io/v1/subtitle-eraser/jobs"
headers = {"X-API-Key": "YOUR_API_KEY"}
payload = {"input": "s3://bucket/episode.mp4", "regions": "auto", "lang": "zh"}

resp = requests.post(API, json=payload, headers=headers)
job = resp.json()
while job["status"] not in ("succeeded", "failed"):
    time.sleep(3)
    job = requests.get(job["url"], headers=headers).json()

print("Output:", job["output_url"])

Pro tips

• Clean master once, dub N times — never re-inpaint per language.
• QC regions on the first episode of each new show before batch processing.
• Use the detected-regions sidecar to auto-skip already-clean scenes.

Run the Subtitle Eraser collection — short-drama sample prefilled.