Whisper AI Service API Documentation

Welcome to the API documentation for the Whisper AI Transcription Service. This service offers a simple and powerful way to transcribe audio files to text using OpenAI Whisper models.

Version: 1.0.0
Host: https://transcription.wraiter.it
Base Model: Whisper AI (base)

Overview

The Whisper AI Transcription Service is a FastAPI-based microservice that uses OpenAI Whisper to transcribe audio files to text. The service accepts audio files in various formats and can automatically recognize the language of the content, with optimized support for Italian.

This API is designed to be intuitive and easy to integrate into any application that requires audio transcription functionality.

Authentication

Currently, the service does not require authentication and is publicly accessible. However, it is subject to rate limiting to prevent abuse.

Rate Limiting

To ensure service availability for all users, the following usage limits are in place:

  • Maximum 10 requests per minute per IP
  • Maximum file size: 25MB
  • Maximum audio duration: 10 minutes

Error Codes

The API uses standard HTTP status codes to indicate the success or failure of a request. Here are the main error codes you might encounter:

Code Description
200 Success. The request has been processed correctly.
400 Bad request. May be due to missing parameters or invalid files.
500 Internal server error. A problem occurred while processing the request.

API Reference

This section describes in detail the endpoints available in the API and how to use them.

POST /transcribe/

This endpoint accepts an audio file and transcribes it to text. It supports various audio formats such as MP3, WAV, M4A, FLAC, and others.

Request Parameters
Name Type Required Description
audio_file File Yes The audio file to transcribe. Must be a valid audio format.
language String No Language code of the audio (e.g., "it" for Italian). If not specified, Whisper will attempt to automatically detect the language.
Responses
200 OK

The transcription was completed successfully.

{
  "transcription": "Text transcribed from the audio file."
}
400 Bad Request

Missing or invalid parameters.

{
  "detail": "The uploaded file is not an audio file"
}
500 Internal Server Error

An error occurred while processing the request.

{
  "detail": "Error during transcription: [error message]"
}

Response Format

All responses are in JSON format and include the following fields:

Field Type Description
transcription String The text transcribed from the audio file. 
{
  "transcription": "This is an example of text transcribed from an audio file."
}

Request Examples

Example with cURL 
curl -X POST "https://transcription.wraiter.it/transcribe/" \
  -H "accept: application/json" \
  -H "Content-Type: multipart/form-data" \
  -F "audio_file=@/path/to/audio_file.mp3" \
  -F "language=it"
Example with JavaScript (Fetch API) 
const formData = new FormData();
formData.append('audio_file', fileInput.files[0]);
formData.append('language', 'it');

fetch('https://transcription.wraiter.it/transcribe/', {
  method: 'POST',
  body: formData
})
.then(response => response.json())
.then(data => {
  console.log('Transcription:', data.transcription);
})
.catch(error => {
  console.error('Error:', error);
});
Example with Python (requests) 
import requests

url = "https://transcription.wraiter.it/transcribe/"

files = {
    'audio_file': open('audio_file.mp3', 'rb')
}

data = {
    'language': 'it'
}

response = requests.post(url, files=files, data=data)
result = response.json()

print("Transcription:", result['transcription'])

Contact

For questions, support, or feedback about this API, you can contact us through the following channels: