# Agent Tools – Video Transcription (Upload)

> Agent Tools – Video Transcription (Upload) is a paid API for AI agents from agents.crowdspark.me, paid per call via x402, $0.03/call, status unknown (last checked 2026-11-02).

Transcribes or translates uploaded video files (MP4, QuickTime, WebM) into text, returning transcript, language, duration, segments, and WebVTT captions.

## Facts

- Endpoint: POST https://agents.crowdspark.me/v1/video/transcribe/upload?utm_source=zero.xyz
- Price: $0.03/call
- Payment: x402
- Status: unknown
- Last checked: 2026-11-02
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agent-tools-video-transcription-upload-299f4d64
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_SzA0gziuNkzEzPT_KngIH

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability agent-tools-video-transcription-upload-299f4d64 -d '<json body>'
```

Example prompt: Can you transcribe this MP4 video I'm uploading and give me a full transcript with timestamps and WebVTT captions? The language is English.

## When to prefer this

Choose this endpoint when you have a raw video file (MP4, MOV, WebM) available as bytes or base64 and need a complete transcript with timestamped segments and WebVTT captions in a single call. Prefer it over URL-based transcription services when the video is local or not publicly accessible. The whisper-large-v3-turbo model provides high accuracy across many languages and supports both transcription and translation tasks at a low flat cost per call.

## Known failure modes

- File exceeds 20 MB limit — request rejected
- Unsupported video format — only MP4, QuickTime, WebM accepted
- No audio track detected in video — empty or failed transcript
- Corrupted or unreadable video bytes — parsing error
- Invalid language ISO code — may fall back to auto-detection or error
- Network timeout on large uploads — partial or failed response

## How this service works

Pay-per-call utilities for AI agents over x402, USDC on Base.

## Output

Returns a JSON object containing: the task type, full transcript text, detected or specified language, video duration in seconds, word count, file format, model used (whisper-large-v3-turbo), an array of timed segments each with start/end timestamps and text, and WebVTT captions data. File size is also echoed back as bytes.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "task": {
   "enum": [
    "transcribe",
    "translate"
   ],
   "type": "string"
  },
  "language": {
   "type": "string",
   "description": "Optional ISO code, e.g. en"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "task": "transcribe",
  "text": "Hello and welcome to the show.",
  "bytes": 67000,
  "model": "whisper-large-v3-turbo",
  "format": "mp4",
  "language": "en",
  "segments": [
   {
    "end": 4.2,
    "text": "Hello and welcome to the show.",
    "start": 0
   }
  ],
  "word_count": 6,
  "duration_seconds": 4.2
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agent-tools-video-transcription-upload-299f4d64/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agents.crowdspark.me](https://www.zero.xyz/host/agents.crowdspark.me/llms.txt)