Speech-to-text (OpenAI transcription wire)

OpenAI's own transcription wire: POST multipart/form-data with a `file` part and get the transcript back. Point any Whisper-shaped SDK at this gateway and pay per call with USDC, no account and no API key. Same model and same four-minute cap as /api/transcribe, which takes a URL instead of an upload; /v1/pro/audio/transcriptions takes it to ten minutes.

TRY IN PLAYGROUND →

Input

Example output

{
  "text": "Example transcript.",
  "duration": 3.2,
  "model": "gpt-transcribe"
}

Try it - see the 402 challenge (free)

curl -i -X POST https://agent402.tools/v1/audio/transcriptions \
  -H "Content-Type: application/json" \
  -d '{"file":"<audio bytes, multipart part named file>","language":"en"}'

The response is HTTP 402 Payment Required with exact payment requirements. Any x402 v2 client pays automatically and retries:

Paid call (JavaScript agent)

import { wrapFetchWithPayment } from "@x402/fetch";
import { x402Client } from "@x402/core/client";
import { registerExactEvmScheme } from "@x402/evm/exact/client";
import { privateKeyToAccount } from "viem/accounts";
import { metaTitle, metaDescription } from "./seo-meta.js";

const client = new x402Client();
client.setSpendControls?.(false); // @x402/core 2.23+ refuses anything over $1 or off the pegged-asset list by default; keep your own ceiling in code instead
registerExactEvmScheme(client, { signer: privateKeyToAccount(KEY) });
const payFetch = wrapFetchWithPayment(fetch, client);

const res = await payFetch("https://agent402.tools/v1/audio/transcriptions", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({
    "file": "<audio bytes, multipart part named file>",
    "language": "en"
  }),
});

Related tools