Video Caption MCP
Generate accurate captions and frame-level descriptions for video content using AI. Returns timestamped caption tracks (SRT/VTT), scene-by-scene narrative descriptions, and key visual event logs. Supports MP4, MOV, and AVI.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
video_url |
string | ✅ | Video URL or local path |
caption_type |
string | ❌ | speech, visual, or both (default: both) |
output_format |
string | ❌ | srt, vtt, json (default: json) |
frame_interval |
float | ❌ | Seconds between frame descriptions (default: 5.0) |
Example Response
{
"duration_seconds": 124.5,
"speech_captions_count": 38,
"visual_events": [
{"timestamp": "00:00:03", "description": "Speaker appears at podium, conference hall background"},
{"timestamp": "00:00:15", "description": "Slide shown: 'Q4 Revenue Results — $2.4B'"},
{"timestamp": "00:01:42", "description": "Audience applauding, speaker gestures to screen"}
],
"output_srt_url": "/output/captions_20240408.srt",
"language_detected": "en"
}
Use Cases
- Corporate training video ADA compliance captioning
- Social media video content description for visually impaired
- Legal evidence video transcription and annotation
Access
# x402 endpoint — pay $0.05 USDC per call (Base mainnet)
POST https://x402.ntriq.co.kr/video-caption
# Service catalog
curl https://x402.ntriq.co.kr/services
x402 micropayments — USDC on Base, gasless EIP-3009