Whisper语音转文字

whisper Model API Reference

This API is based on the Whisper model and provides speech-to-text functionality, supporting common audio formats.

Basic Concepts

• Whisper model: OpenAI’s open-source speech recognition model, supporting multilingual transcription
• Audio formats: Supports common formats such as mp3, wav, and m4a

API Endpoint

POST https://api.4allapi.com/v1/audio/transcriptions

Request Parameters

Parameter	Type	Required	Description
model	string	Yes	Fixed value "whisper-1"
file	file	Yes	The audio file to transcribe

Request Headers

Authorization: Bearer sk-*********************  # replace with your API token

Python Example

import json
import requests

def voice_to_text(file_path):
    """
    Speech-to-text functionality

    Parameters:
    file_path: path to the audio file

    Returns:
    recognized text content
    """
    url = "https://api.4allapi.com/v1/audio/transcriptions"

    # Build request parameters
    payload = {"model": "whisper-1"}
    files = {"file": ("audio.mp3", open(file_path, "rb"))}

    # Set request headers (replace with your API key)
    headers = {"Authorization": "Bearer sk-***************************"}  # replace with your API token

    # Send POST request
    response = requests.post(url, headers=headers, data=payload, files=files)

    # Parse response data
    data = json.loads(response.text)

    # Return the transcription result
    return data.get("text", "")

# Example usage
print(voice_to_text("audio.mp3"))  # replace with the path to your audio file

Response Example

Successful response:

{
    "text": "This is the recognized text content"
}

Notes

1. It is recommended that the audio file size not exceed 25MB
2. Supports multiple languages, including Chinese and English
3. Please keep your API key secure and do not disclose it

---

4All API Footer

4All API · One-stop AI foundation model API aggregation platform | Pricing | Contact Us

© 2025 4All API. All rights reserved.