# VoxSift Transcription API

> VoxSift Transcription API is a paid API for AI agents from voxsift-x402.fly.dev, paid per call via x402, $0.2/call, status unknown (last checked 2026-09-15).

Transcribes a public HTTPS audio or video URL into structured JSON with timestamps, optional speaker diarization, and confidence metadata.

## Facts

- Endpoint: POST https://voxsift-x402.fly.dev/v1/transcribe
- Price: $0.2/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/voxsift-transcription-api-7cf3909c
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_V0WbEhh2kUvdyXdYhv_l5

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 voxsift-transcription-api-7cf3909c -d '<json body>'
```

Example prompt: Transcribe this podcast episode for me — https://example.com/podcast/ep42.mp3 — with speaker diarization turned on and language set to English (en).

## When to prefer this

Choose VoxSift when you need structured JSON transcription output with per-utterance timestamps and confidence metadata from a publicly accessible audio or video URL, especially when speaker diarization is required. Ideal for automating podcast indexing, meeting summarization, accessibility captioning pipelines, or any workflow where machine-readable timestamped speech output is needed. Prefer this over generic STT APIs when you want actual-duration-based billing and speaker-separated utterances in a single call.

## Known failure modes

- URL is not publicly accessible or requires authentication — returns an error indicating the resource could not be fetched
- Non-audio/video URL or unsupported media format — returns format error
- Network timeout fetching long media files — may return a timeout or partial result error
- Unsupported or unrecognized language code — may default to auto-detect or return a validation error
- Audio too short or silent — may return an empty transcript or low-confidence result

## How this service works

Metered speaker-aware audio transcription for autonomous agents. Submit a public HTTPS audio or video URL and receive deterministic transcript JSON with timestamps, optional speaker diarization, confidence metadata, and actual-duration x402 settlement.

## Output

Returns structured JSON containing the full transcript broken into utterances, each with start/end timestamps, speaker identifiers (when diarization is enabled), confidence scores per segment, and the actual audio duration for billing settlement purposes.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "format": "uri"
  },
  "diarize": {
   "type": "boolean",
   "default": true
  },
  "language": {
   "type": "string"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "text": "Example transcript text.",
  "billing": {
   "scheme": "x402-upto",
   "charged_usd": "$0.002084",
   "charged_atomic": "2084"
  },
  "language": "en",
  "segments": [
   {
    "end": 4.2,
    "text": "Example transcript text.",
    "start": 0,
    "speaker": "A",
    "confidence": 0.96
   }
  ],
  "confidence": 0.96,
  "request_id": "3d0ee76d-8337-41c6-bf44-e1eb1547f249",
  "duration_seconds": 12.5,
  "source_duration_seconds": 12.5
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/voxsift-transcription-api-7cf3909c/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from voxsift-x402.fly.dev](https://www.zero.xyz/host/voxsift-x402.fly.dev/llms.txt)