# NoJunk Speech – Audio Transcription

> NoJunk Speech – Audio Transcription is a paid API for AI agents from nojunk-speech.mick8511.workers.dev, paid per call via x402, $0.02/call, status unknown (last checked 2026-10-29).

Transcribes an audio file (supplied as a public URL or base64-encoded bytes) to timestamped text, detecting the spoken language automatically.

## Facts

- Endpoint: POST https://nojunk-speech.mick8511.workers.dev/transcribe?utm_source=zero.xyz
- Price: $0.02/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-29
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/nojunk-speech-audio-transcription-a988c185
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_Sx3ie-46EF3zvFxtmzLBy

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 nojunk-speech-audio-transcription-a988c185 -d '<json body>'
```

Example prompt: Can you transcribe the audio at https://example.com/interview.mp3 — I need the full text with timestamps and want to know what language it's in?

## When to prefer this

Choose this endpoint when you need pay-per-use speech-to-text with no subscription or quota management, especially for AI agent workflows where cost should only be incurred on success. Ideal when you need timestamped segments alongside the transcript, automatic language detection, or when submitting audio by public URL rather than a streaming API. Compared to batch transcription services, this returns results synchronously per call, making it suitable for real-time or single-file agent tasks up to ~10 minutes long.

## Known failure modes

- Audio URL not publicly accessible or returns non-2xx — request fails with an error
- Audio file exceeds 20 MB limit — rejected before transcription
- Unsupported audio format provided — conversion error returned
- Both audio_url and audio_base64 provided or neither provided — validation error
- Low-quality or highly noisy audio may produce inaccurate transcripts with low language_probability
- Payment not settled via x402/USDC — call not executed and no charge incurred

## How this service works

NoJunk Speech transcribe: Transcribe audio (URL or base64, up to 20 MB / about 10 minutes) to text with timestamps, word count and detected language (Whisper large-v3-turbo).

## Output

Returns a JSON object containing the full transcript text, the detected ISO 639-1 language code and confidence probability, an array of timestamped segments (each with start time, end time, and segment text), total word count, audio duration in seconds, file size in bytes, the Whisper model used, and a generated_at timestamp.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "language": {
   "type": "string",
   "description": "Optional ISO 639-1 language hint such as \"en\"; omit to auto-detect"
  },
  "audio_url": {
   "type": "string",
   "description": "Public http(s) URL of an audio file (mp3, wav, m4a, ogg, flac, webm, mp4), max 20 MB; use this OR audio_base64"
  },
  "audio_base64": {
   "type": "string",
   "description": "Base64-encoded audio bytes (max 20 MB decoded); use this OR audio_url"
  },
  "content_type": {
   "type": "string",
   "description": "Optional media type of audio_base64, e.g. audio/mpeg"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "text": "Welcome back to the show. Today we are talking about how AI agents pay for the tools they use.",
  "model": "@cf/openai/whisper-large-v3-turbo",
  "source": "url",
  "language": "en",
  "segments": [
   {
    "end": 2.1,
    "text": "Welcome back to the show.",
    "start": 0
   },
   {
    "end": 6.4,
    "text": "Today we are talking about how AI agents pay for the tools they use.",
    "start": 2.1
   }
  ],
  "word_count": 18,
  "audio_bytes": 102400,
  "generated_at": "2026-10-15T12:00:00.000Z",
  "duration_seconds": 6.4,
  "language_probability": 0.99
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/nojunk-speech-audio-transcription-a988c185/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from nojunk-speech.mick8511.workers.dev](https://www.zero.xyz/host/nojunk-speech.mick8511.workers.dev/llms.txt)