# 2s Audio Transcription API

> 2s Audio Transcription API is a paid API for AI agents from 2s.io, paid per call via x402, $0.1305/call, status unknown (last checked 2026-09-15).

Transcribes an audio file from a public URL into text using Deepgram nova-2, with optional speaker diarization and language detection

## Facts

- Endpoint: POST https://2s.io/api/transcribe/audio
- Price: $0.1305/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Success rate: 0% of calls made through Zero
- Activations on Zero: 2
- Tags: x402
- Canonical page: https://www.zero.xyz/c/2s-audio-transcription-api-1c795df2
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_rbyVKbczp7UZEPIiK8L5n

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 2s-audio-transcription-api-1c795df2 -d '<json body>'
```

Example prompt: Can you transcribe this audio file for me — https://example.com/interview.mp3 — and label which speaker is saying what throughout the recording?

## When to prefer this

Choose this endpoint when you need pay-per-call audio transcription without API key setup or subscriptions, especially in agentic workflows that pay via USDC on Base or Solana. It is ideal for one-off transcription tasks, pipelines with variable volume, or when speaker diarization is needed alongside transcription. Prefer this over self-hosted or subscription Deepgram access when zero-setup and crypto-native payment are priorities.

## Known failure modes

- Audio file exceeds 15 MB or 15 minutes — server rejects the request
- URL is not publicly accessible or returns a non-200 response
- Unsupported audio format causes decoding failure
- Invalid BCP-47 language code causes a validation error
- Payment failure via x402 results in 402 response with no transcription
- Network timeout if audio download takes too long

## How this service works

The (most) everything API: 575+ pay-per-call endpoints for AI agents — ground-truth data, a full AI gateway, and agent infrastructure (storage, queues, watchers). USDC via x402, no signup, no API keys, and upto billing: pay actual usage, not the quote.

## Output

A JSON object containing the full transcript text, word-level tokens with start/end timestamps and confidence scores, detected language, total duration in seconds, word count, and optionally labeled speaker utterances. Also includes source attribution indicating Deepgram nova-2 as the underlying provider.

## Example request

```json
{
 "url": "https://www.soundhelix.com/examples/mp3/SoundHelix-Song-1.mp3",
 "diarize": false,
 "language": "en"
}
```

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "description": "Public URL of the audio file (≤15 MB, ≤15 min)."
  },
  "diarize": {
   "type": "boolean",
   "default": false,
   "description": "Label speakers and return utterance segments."
  },
  "language": {
   "type": "string",
   "description": "BCP-47 language hint, e.g. \"en\", \"es\". Auto-detected when omitted."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "schema": {
  "type": "object",
  "properties": {
   "ok": {
    "type": "boolean"
   },
   "meta": {
    "type": "object",
    "properties": {
     "wordCount": {
      "type": "integer"
     }
    }
   },
   "items": {
    "type": "array",
    "items": {
     "type": "object",
     "properties": {
      "words": {
       "type": "array",
       "items": {
        "type": "object",
        "properties": {
         "end": {
          "type": "number"
         },
         "word": {
          "type": "string"
         },
         "start": {
          "type": "number"
         },
         "speaker": {},
         "confidence": {
          "type": "number"
         }
        }
       }
      },
      "language": {
       "type": "string"
      },
      "confidence": {
       "type": "number"
      },
      "transcript": {
       "type": "string"
      },
      "utterances": {
       "type": "array",
       "items": {}
      },
      "durationSeconds": {
       "type": "number"
      }
     }
    }
   },
   "total": {
    "type": "integer"
   },
   "source": {
    "type": "object",
    "properties": {
     "url": {
      "type": "string"
     },
     "license": {
      "type": "string"
     },
     "provider": {
      "type": "string"
     }
    }
   }
  }
 },
 "example": {
  "ok": true,
  "meta": {
   "wordCount": 52
  },
  "items": [
   {
    "words": [
     {
      "end": 0.24,
      "word": "as",
      "start": 0.08,
      "speaker": null,
      "confidence": 0.99
     }
    ],
    "language": "en",
    "confidence": 0.998,
    "transcript": "As much as it’s worth celebrating the first spacewalk with an all-female team…",
    "utterances": [],
    "durationSeconds": 25.9
   }
  ],
  "total": 1,
  "source": {
   "url": "https://deepgram.com",
   "license": "Generated transcript — Deepgram API terms",
   "provider": "Deepgram (nova-2)"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/2s-audio-transcription-api-1c795df2/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from 2s.io](https://www.zero.xyz/host/2s.io/llms.txt)