UnmarkAIunmarkai

Built for developers

Video Text Removal API & Subtitle Remover

Remove burned-in subtitles, captions, and text overlays at scale. Upload a video or pass a URL. Get a clean result, ready for new copy.

  • File & URL input
  • 3 removal modes
  • Signed webhooks
Get API keyRead API docsUse this n8n workflow

$1 API trial credit for eligible new users. Trial videos include an UnmarkAI watermark. Top up your API wallet for watermark-free results.See API pricing →

See the difference

Two-line captions. Clean video.

Actual API result
BeforeBurned-in captions
Tutorial video with two lines of bold captions and a yellow highlighted word
AfterText removed
The same tutorial frame after the API removed the two-line captions
Video length
12 seconds
Removal mode
Selected area
Recorded charge
$0.045
Watch the 12-second result

Real API results · short video examples

Compare six cases: tutorial captions, product copy, vertical creator captions, title text, a sticker label and a security-cam timestamp. Each 12-second comparison keeps the input and actual API output synchronized.

Remove text from your video →Build with the API →

Measured 2026-10-06 · USD API wallet; web points are separate. Runtime depends on the clip and queue. See case details for the billable seconds recorded for each run.

Demo overlays added to licensed footage; outputs come from the real API with no manual cleanup. These are examples, not a quality or speed guarantee.

Licensed footage: Sururi Ballıdağ Director · Tima Miroshnichenko · ANTONI SHKRABA production · Get Lost Mike · Tima Miroshnichenko · Pressmaster

Tutorial captions →UGC product copy →Vertical creator captions →Title and headline text →Sticker and label text →Date and time stamp →

One durable job lifecycle

For individual uploads, remove text from video online and compare the preview.

From captioned source to clean video

Remove burned-in subtitles without proxying large files through your application server. Each upload, job, and result has a stable identifier that your backend can store and reconcile.

  1. 01POST /v1/uploads or /v1/url-uploads

    Choose local upload or URL import

    Request a one-time signed upload for local bytes, or asynchronously import a directly downloadable public HTTPS video URL.

  2. 02PUT signed URL

    Upload video bytes directly

    Send the source file to object storage without proxying a multi-gigabyte video through your application server.

  3. 03POST /v1/jobs

    Create an idempotent job

    Choose full-frame, selected-area or multi-region cleanup and attach your own metadata for order, asset, or tenant reconciliation.

  4. 04GET /v1/jobs/{id}

    Receive the clean result

    Follow progress with the Jobs API or signed webhook events, then download the result from its temporary signed URL.

Three cleanup modes

Choose how the API removes subtitles and text

Use the same asynchronous job contract for broad discovery or precise, predictable regions. The mode is explicit in every request and response.

all_area

Full-frame subtitle and text removal

Scan the full video frame when captions, labels, or overlays can appear in different positions. No rectangle is supplied.

sel_area

Selected-area subtitle removal

Provide source-video pixel coordinates when a timestamp, lower third, or caption band stays inside a known region.

multi_area

Multiple text regions in one job

Send an array of non-overlapping source-pixel rectangles to clear separated labels or caption bands. Billed at the same rate as all_area. See the two-region product example above.

Top up over time. Pay less per minute.

API base prices are now 25% lower. Save up to a further 40% with API wallet tiers. Your tier is based on net API wallet top-ups in the past 365 days; qualifying payments can add up across multiple top-ups.

At $500 in qualifying top-ups: $0.18/minSelected-area removal · 20% off the base rate
USD per minute at each qualifying top-up tier
Net top-ups in past 365 daysDiscountSelected areaFull frame
Base rate—$0.225$0.45
$100+10% off$0.2025$0.405
$500+20% off$0.18$0.36
$2,000+25% off$0.16878$0.3375
$5,000+30% off$0.1575$0.315
$10,000+35% off$0.14628$0.2925
$20,000+40% off$0.135$0.27

The lowest rate requires $20,000 in qualifying net top-ups. Each payment counts for 365 days from settlement; refunds, chargebacks and aging payments can lower your tier. This is a usage-rate discount, not a discount on the amount deposited. Promotional credits do not qualify.

API wallet funds do not expire. The discount applies to both text removal and video translation; website points and subscriptions are separate. Jobs lock their rate when funds are reserved. Per-minute figures are equivalents of per-second billing, rounded up to whole seconds with a five-second minimum; clips under five seconds are billed as five seconds. Pricing rules · Check your wallet tier. GET /v1/balance provides your authoritative effective rates.

Compare video text removal APIs

Compare published API rates and documented inputs for authorized video cleanup. Selected-area removal and automatic full-frame cleanup are different workflows; test the same footage before comparing output quality.

USD API rates · Sources checked September 8, 2026
Provider / modePublished rateFeatures and billing conditions
UnmarkAI · selected area$0.225/min base
$0.18/min at $500 qualifying top-ups
$0.135/min at $20,000		Rectangle coordinates target the caption region. 15-second minimum; clips under 15 seconds are billed as 15 seconds. Discounts use trailing-365-day net top-ups. Official pricing.
UnmarkAI · full frame$0.45/min base
$0.36/min at $500 qualifying top-ups
$0.27/min at $20,000		Whole-frame text cleanup. Local uploads or HTTPS imports, asynchronous jobs and signed webhooks. Same tier rules. API contract.
WaveSpeed · Video Watermark Remover$0.010/sec
$0.60/min equivalent		Video URL input; advertised for captions, logos and text. Documentation lists a three-second minimum and up to 10-minute clips. Rate derived from its $0.05/5-sec table; final task charge prevails. Official model pricing.
MuAPI · AI Video Watermark Remover$0.065 for ≤5 sec;
$0.013/sec beyond that length		Advertised for logos, captions and unwanted text. This is the general AI Video Watermark Remover; its separate Seedance 2 remover is listed at $0.025 for ≤5 sec, then $0.005/sec. Confirm which model fits your source. Official model rates.
VModel · Video Watermark Remover$0.020/sec
$1.20/min equivalent		Video plus a mask image input. Public model rate is per input-video second; contact the provider for volume pricing. Official model pricing.

UnmarkAI full-frame base pricing is $0.45/min after the 25% price reduction; qualifying top-ups lower it to $0.36/min at the $500 tier. Selected-area UnmarkAI pricing is lower, but requires a region. Competitor private discounts, taxes and output quality are not normalized here; this is not a measured quality ranking.

For an effect comparison, inspect text remnants, moving backgrounds, faces crossing the caption region, temporal flicker and original resolution. No side-by-side benchmark is claimed.

Production controls

Designed for automation that has to recover

The API exposes the state and identifiers your workers need to retry safely, trace failures, and keep customer workflows moving.

01

Safe retries

A required Idempotency-Key protects job creation from duplicate charges when networks fail or workers retry.

02

Asynchronous by design

Queue larger videos without holding an HTTP request open. Read authoritative status and smooth progress from the Job resource.

03

Signed event delivery

Verify one HTTPS webhook endpoint and receive state changes on your server while retaining polling as a recovery path.

04

Account-level controls

Create scoped keys, revoke credentials, review wallet activity, and inspect live rates from the API Console.

Server-side quickstart

Run your first subtitle removal API job

Create a key with jobs:write and jobs:read, fund the API wallet, and save the key as an environment variable. With Bash, cURL, and jq installed, this example creates a signed upload, transfers the video, starts an idempotent job, and checks its status once.

  • Base URL: https://api.unmarkai.net/v1
  • Authentication: Bearer uma_live_*
  • Job type: remove_text
Open the complete API quickstart

Python & cURL examples on GitHub

cURL + jq · bash
export UNMARKAI_API_KEY='uma_live_your_key_here'
export IDEMPOTENCY_KEY="remove-$(date +%s)"
API_BASE='https://api.unmarkai.net/v1'
VIDEO='./source.mp4'
SIZE=$(wc -c < "$VIDEO" | tr -d ' ')

# 1. Create a one-time upload ticket.
curl --fail-with-body -sS -X POST "$API_BASE/uploads" \
  -H "Authorization: Bearer $UNMARKAI_API_KEY" \
  -H 'Content-Type: application/json' \
  --data "{"filename":"source.mp4","content_type":"video/mp4","file_size_bytes":$SIZE}" \
  -o ticket.json

# 2. Upload the bytes with every signed header unchanged.
jq -r '.headers | to_entries[] | "\(.key): \(.value)"' ticket.json > upload-headers.txt
curl --fail-with-body -sS -X PUT "$(jq -r '.url' ticket.json)" \
  --header '@upload-headers.txt' --upload-file "$VIDEO"

# 3. Create an asynchronous, idempotent cleanup job.
UPLOAD_ID=$(jq -r '.upload_id' ticket.json)
curl --fail-with-body -sS -X POST "$API_BASE/jobs" \
  -H "Authorization: Bearer $UNMARKAI_API_KEY" \
  -H 'Content-Type: application/json' \
  -H "Idempotency-Key: $IDEMPOTENCY_KEY" \
  --data "{"type":"remove_text","input":{"upload_id":"$UPLOAD_ID"},"mode":"all_area","rect":null}" \
  -o job.json

# 4. Check status once. Poll this resource, or receive signed webhook events.
curl --fail-with-body -sS "$API_BASE/jobs/$(jq -r '.id' job.json)" \
  -H "Authorization: Bearer $UNMARKAI_API_KEY" | jq .

Clear operating boundaries

Usage-based processing with server-side security

Videos can be up to 2 GiB and 60 minutes under the default public contract. Jobs reserve an estimate, settle against measured billable video seconds, and release unused funds after completion or failure.

Authoritative rates
GET /v1/balance
Concurrency
4 base · up to 10 with capacity packs
Maximum video
2 GiB · 60 minutes
Credential boundary
Trusted server only
Keep API keys out of client code.

Store uma_live_* credentials in environment variables or a secret manager. Use scoped keys per environment, rotate them when access changes, and retain response request IDs for support and tracing.

Native MCP endpoint

Give your AI agent the cleanup tools

The same engine is also a hosted MCP server at https://api.unmarkai.net/mcp. One uma_live_* key with the same scopes unlocks nine tools for Claude Code, Cursor, Codex, opencode, and any MCP-capable agent — no SDK, no server to run. Agents can upload local files with curl or import a directly downloadable HTTPS video URL asynchronously.

  • create_upload

    Mint a one-time signed upload ticket for a local video file.

  • create_url_upload

    Import a directly downloadable HTTPS video URL asynchronously.

  • get_upload

    Wait for a URL import to become uploaded, failed, or expired.

  • remove_text

    Start an idempotent cleanup job in all_area, sel_area, or multi_area mode.

  • get_job

    Read progress; optionally wait server-side for a terminal status.

  • list_jobs

    Page through recent jobs filtered by status.

  • cancel_job

    Stop a validating or queued job before processing starts.

  • get_balance

    Check the API wallet, live rates, and tier multipliers.

  • estimate_cost

    Preview billable seconds and USD cost before spending.

Get the agent skills on GitHub
Claude Code · Cursor
# Claude Code
claude mcp add --transport http unmarkai \
  https://api.unmarkai.net/mcp \
  --header "Authorization: Bearer $UNMARKAI_API_KEY"

# Cursor · ~/.cursor/mcp.json
{
  "mcpServers": {
    "unmarkai": {
      "type": "http",
      "url": "https://api.unmarkai.net/mcp",
      "headers": { "Authorization": "Bearer $UNMARKAI_API_KEY" }
    }
  }
}

Integration questions

Before you automate video cleanup

Use the API only with video you own, license, produce for a client, or otherwise have permission to modify. Preserve required disclosures, attribution, and legal notices.

Can I try the API for free, and will results have a watermark?

Eligible new users receive $1 in API trial credit. Free trial videos include an UnmarkAI watermark. A successful API wallet top-up or an admin API balance grant unlocks watermark-free final videos. Web points and subscriptions do not qualify. Existing accounts do not receive a retroactive trial grant.

What is the Video Subtitle Remover API?

The Video Subtitle Remover API removes subtitles and captions that are already rendered into the video pixels, then reconstructs the selected area or full frame. It is designed for authorized localization, caption replacement, and video-library cleanup workflows.

Is this API for burned-in subtitles or selectable subtitle tracks?

Use this API for hardcoded or burned-in subtitles that remain visible after subtitle tracks are disabled. If a video contains a separate SRT, VTT, or ASS track, remove or replace that track directly instead of reconstructing video pixels.

Can the API remove two-line captions with a highlighted word?

Yes. A remove_text job can process burned-in two-line captions whose active word changes color and bounces. Use sel_area when both lines stay inside one region, covering the largest word expansion and outline; use all_area when the caption block moves across the frame. The 12-second tutorial and vertical creator cases above show actual API outputs, exact settings and recorded charges. Inspect every word change for remnants and flicker.

Can Claude, Cursor, or Codex call this API for me?

Yes. Point any MCP-capable agent at the hosted MCP server, https://api.unmarkai.net/mcp, with the same uma_live_* key and scopes. Nine tools cover local uploads, direct HTTPS URL imports, job creation, status, cancellation, balance, and cost estimates, and the open-source agent skills on GitHub teach both input flows. Local file transfer stays in your shell; URL imports run asynchronously.

Can the API also remove timestamps and other video text?

Yes. The same remove_text job can clean timestamps, labels, lower thirds, and other static text overlays. Results depend on motion, background detail, text size, and how much of the scene is covered.

Which removal mode should I use?

Use all_area when text can appear across the frame. Use sel_area for one region covering at most half of the source frame. Use multi_area with an array of source-pixel rectangles for multiple regions, billed at the same rate as all_area.

Can I call the Video Text Removal API from a browser?

Do not embed an uma_live_* key in browser JavaScript, mobile apps, URLs, logs, or public repositories. Call the API from a trusted server, worker, script, or automation environment and keep the key in a secret manager.

How is API usage billed?

The API uses a separate USD wallet and charges by billable video seconds, rounded up to whole seconds with a 15-second minimum. Clips under 15 seconds are billed as 15 seconds. The effective rate depends on cleanup mode and account tier. Query GET /v1/balance for the authoritative live rates before showing estimates to customers.

Translate after removing the old subtitles

Use the clean video in a separate Video Translation & AI Dubbing API job to add translated voiceover and captions, with optional text review.

Build with UnMarkAI

Add subtitle removal to your video pipeline

Create a server-side key, check the live usage rate, and follow the quickstart from captioned source to clean result.

Open API Console Remove one video's subtitles online