Sign in Sign up

Welcome

AudioTranscribe API Documentation

Overview

The AudioTranscribe API allows you to stream audio data via WebSocket and receive real-time voice transcriptions. This service is ideal for applications requiring live speech-to-text functionality.

WebSocket Endpoint

wss://api.audiotranscribe.com/v1/transcribe

Authentication

To use the AudioTranscribe API, you need to include your API key in the WebSocket connection request headers:

Authorization: Bearer YOUR_API_KEY

Streaming Protocol

  1. Establish a WebSocket connection to the endpoint.
  2. Send audio data in binary format.
  3. Receive transcription results as they become available.

Audio Format

  • Supported audio format: 16-bit PCM
  • Sample rate: 16000 Hz
  • Channels: Mono

Message Format

Client to Server

Send raw audio data as binary WebSocket messages.

Server to Client

Transcription results are sent as JSON objects:

{
  "transcript": "This is the transcribed text.",
  "confidence": 0.95,
  "is_final": false
}
  • transcript: The transcribed text
  • confidence: Confidence score between 0 and 1
  • is_final: Boolean indicating if this is the final result for the current audio segment

Example Usage

Here's a simple example using JavaScript and the WebSocket API:

const socket = new WebSocket('wss://api.audiotranscribe.com/v1/transcribe');

socket.onopen = function(event) {
  console.log('WebSocket connection established');
  // Start sending audio data
};

socket.onmessage = function(event) {
  const result = JSON.parse(event.data);
  console.log('Transcription:', result.transcript);
};

// Function to send audio data (implement this based on your audio source)
function sendAudioData(audioChunk) {
  if (socket.readyState === WebSocket.OPEN) {
    socket.send(audioChunk);
  }
}

Error Handling

The server may send error messages in the following format:

{
  "error": "Error message description"
}

Common errors include authentication failures, invalid audio format, or service unavailability.

Rate Limiting

  • Free tier: 10 minutes of audio per day
  • Pro tier: 1000 minutes of audio per day

Exceeding these limits will result in a rate limit error message.

Support

For any issues or questions, please contact our support team at [email protected].