# Whisper: Audio Transcription

Transcribe an audio file (max 10 MB) to text using Cloudflare Workers AI Whisper (model `@cf/openai/whisper`). Returns the full transcription plus optional WebVTT subtitles and per-word timing when supported by the audio.

- Method: POST
- Path: `/transcribe`
- Price: $0.02000000

## Input Schema

```json
{
  "type": "object",
  "properties": {
    "audio_url": {
      "type": "string",
      "format": "uri",
      "description": "Public URL of the audio file to transcribe (max 10 MB)"
    },
    "model": {
      "type": "string",
      "enum": [
        "whisper"
      ],
      "description": "Whisper variant. Currently only `whisper` (@cf/openai/whisper) is supported. Defaults to `whisper`."
    }
  },
  "required": [
    "audio_url"
  ],
  "additionalProperties": false
}
```

## Output Schema

```json
{
  "type": "object",
  "properties": {
    "text": {
      "type": "string",
      "description": "Full transcription"
    },
    "vtt": {
      "type": "string",
      "description": "WebVTT subtitle string with timestamps (when available)"
    },
    "word_count": {
      "type": "integer",
      "minimum": 0
    },
    "words": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "word": {
            "type": "string"
          },
          "start": {
            "type": "number",
            "description": "Start time in seconds"
          },
          "end": {
            "type": "number",
            "description": "End time in seconds"
          }
        },
        "additionalProperties": false
      },
      "description": "Per-word timing (when supported by the model)"
    }
  },
  "required": [
    "text"
  ],
  "additionalProperties": false,
  "description": "Whisper transcription with optional word-level timing"
}
```

## Networks

| Name | Mode | CAIP-2 | Chain ID / Cluster | Protocols |
|---|---|---|---|---|
| Tempo Mainnet | live | `eip155:4217` | 4217 | mpp |
| Solana Mainnet | live | `solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp` | mainnet-beta | mpp, x402 |
| Base | live | `eip155:8453` | 8453 | x402 |

## Accepted Currencies

| Symbol | Name | Decimals | Network | Address | Protocols |
|---|---|---|---|---|---|
| USDC.e | Bridged USDC (Stargate) | 6 | `eip155:4217` | `0x20c000000000000000000000b9537d11c60e8b50` | mpp |
| USDC | USD Coin (Solana) | 6 | `solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp` | `EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v` | mpp, x402 |
| USDC | USD Coin | 6 | `eip155:8453` | `0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913` | x402 |

## How to invoke

Payment is handled automatically — the client signs the 402 challenge and retries.

### TypeScript — Tempo (mppx + fetch) ([docs](https://mpp.dev/sdk/typescript/client))

```ts
import { privateKeyToAccount } from 'viem/accounts'
import { Mppx, tempo } from 'mppx/client'

Mppx.create({
  methods: [tempo({ account: privateKeyToAccount('0x...') })],
})

const res = await fetch('https://cloudflare.payweave.services/transcribe', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    "audio_url": "https://example.com",
    "model": "whisper"
  }),
})
const data = await res.json()
```

### TypeScript — Solana (@solana/mpp + fetch) ([docs](https://github.com/solana-foundation/mpp-sdk))

```ts
import { createKeyPairSignerFromBytes } from '@solana/kit'
import { Mppx } from 'mppx/client'
import { solana } from '@solana/mpp/client'

const signer = await createKeyPairSignerFromBytes(/* 64-byte secret key */)
Mppx.create({ methods: [solana({ signer })] })

const res = await fetch('https://cloudflare.payweave.services/transcribe', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    "audio_url": "https://example.com",
    "model": "whisper"
  }),
})
const data = await res.json()
```

### mppx CLI ([docs](https://mpp.dev/sdk/typescript/cli))

```sh
npx mppx "https://cloudflare.payweave.services/transcribe" -X POST -H 'Content-Type: application/json' -d '{"audio_url":"https://example.com","model":"whisper"}'
```

### Tempo Wallet ([docs](https://docs.tempo.xyz/cli/wallet))

```sh
tempo request "https://cloudflare.payweave.services/transcribe" -X POST --json '{"audio_url":"https://example.com","model":"whisper"}'
```

### pay.sh — Solana Foundation ([docs](https://pay.sh))

```sh
pay --mainnet curl "https://cloudflare.payweave.services/transcribe" -X POST -H 'Content-Type: application/json' -d '{"audio_url":"https://example.com","model":"whisper"}'
```

### agentcash CLI ([docs](https://www.npmjs.com/package/agentcash))

```sh
npx agentcash fetch "https://cloudflare.payweave.services/transcribe" -m POST -b '{"audio_url":"https://example.com","model":"whisper"}' -p mpp
```

### x402 — Base / Solana USDC (x402-fetch) ([docs](https://x402.org))

```ts
import { wrapFetchWithPayment } from 'x402-fetch'
import { privateKeyToAccount } from 'viem/accounts'

const account = privateKeyToAccount('0x...') // Base USDC wallet
const fetchWithPay = wrapFetchWithPayment(fetch, account)

const res = await fetchWithPay('https://cloudflare.payweave.services/transcribe', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    "audio_url": "https://example.com",
    "model": "whisper"
  }),
})
const data = await res.json()
```

### x402 — agentcash CLI

```sh
npx agentcash fetch "https://cloudflare.payweave.services/transcribe" -m POST -b '{"audio_url":"https://example.com","model":"whisper"}' -p x402
```
