Audio to Text

Audio to Text tasks transcribe spoken audio CAPTCHAs into text. Send a base64-encoded audio file and receive the transcribed content. This is commonly used for accessibility-oriented CAPTCHA challenges that read out letters, numbers, or words.

Task Type

Type	Description
AudioToTextTask	Transcribe audio CAPTCHA to text

Parameters

Parameter	Type	Required	Description
type	string	Yes	AudioToTextTask
body	string	Yes	Base64-encoded audio file (MP3, WAV, OGG)
lang	string	No	Language code (e.g., en, ru, de)

Create Task

Request

{
  "clientKey": "YOUR_API_KEY",
  "task": {
    "type": "AudioToTextTask",
    "body": "SUQzBAAAAAAAI1RTU0UAAAA...",
    "lang": "en"
  }
}

Response

{
  "errorId": 0,
  "status": "ready",
  "solution": {
    "text": "seven four two"
  }
}

Solution Object

{
  "text": "seven four two"
}

Field	Type	Description
text	string	The transcribed text from the audio CAPTCHA

Code Examples

Python

import requests
import base64

API_KEY = "YOUR_API_KEY"

# Encode the audio file to base64
with open("captcha.mp3", "rb") as f:
    audio_data = base64.b64encode(f.read()).decode()

# Create task
response = requests.post("https://api.ucaptcha.net/createTask", json={
    "clientKey": API_KEY,
    "task": {
        "type": "AudioToTextTask",
        "body": audio_data,
        "lang": "en"
    }
}).json()

if response.get("status") == "ready":
    print("Transcription:", response["solution"]["text"])
else:
    import time
    task_id = response["taskId"]
    while True:
        result = requests.post("https://api.ucaptcha.net/getTaskResult", json={
            "clientKey": API_KEY,
            "taskId": task_id
        }).json()

        if result["status"] == "ready":
            print("Transcription:", result["solution"]["text"])
            break
        elif result["status"] == "failed":
            print("Error:", result.get("errorDescription"))
            break

        time.sleep(5)

JavaScript

import { readFileSync } from "fs";

const API_KEY = "YOUR_API_KEY";

// Encode the audio file to base64
const audioData = readFileSync("captcha.mp3").toString("base64");

const response = await fetch("https://api.ucaptcha.net/createTask", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({
    clientKey: API_KEY,
    task: {
      type: "AudioToTextTask",
      body: audioData,
      lang: "en"
    }
  })
}).then(r => r.json());

if (response.status === "ready") {
  console.log("Transcription:", response.solution.text);
} else {
  const taskId = response.taskId;
  while (true) {
    const result = await fetch("https://api.ucaptcha.net/getTaskResult", {
      method: "POST",
      headers: { "Content-Type": "application/json" },
      body: JSON.stringify({ clientKey: API_KEY, taskId })
    }).then(r => r.json());

    if (result.status === "ready") {
      console.log("Transcription:", result.solution.text);
      break;
    } else if (result.status === "failed") {
      console.error("Error:", result.errorDescription);
      break;
    }

    await new Promise(r => setTimeout(r, 5000));
  }
}

cURL

# Encode audio to base64
AUDIO_BASE64=$(base64 -w 0 captcha.mp3)

# Create task
curl -X POST https://api.ucaptcha.net/createTask \
  -H "Content-Type: application/json" \
  -d "{
    \"clientKey\": \"YOUR_API_KEY\",
    \"task\": {
      \"type\": \"AudioToTextTask\",
      \"body\": \"$AUDIO_BASE64\",
      \"lang\": \"en\"
    }
  }"

# Poll for result if not immediately ready (replace TASK_ID)
curl -X POST https://api.ucaptcha.net/getTaskResult \
  -H "Content-Type: application/json" \
  -d '{
    "clientKey": "YOUR_API_KEY",
    "taskId": "TASK_ID"
  }'

Tip

Specify the lang parameter when the audio language is known. This improves transcription accuracy, especially for CAPTCHAs that spell out words or use language-specific number pronunciation.

Provider Coverage

• 2Captcha

Aliases

The following legacy type name is also accepted for backward compatibility:

• AudioTask