README.md

---
title: Audio Transcription
emoji: 🎙️
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 4.36.0
app_file: app.py
pinned: false
---

# Multi-Source Audio Transcription with Faster Whisper

This application transcribes audio from multiple sources using Faster Whisper v3 turbo int8, providing a flexible and powerful transcription solution.

## Features

- Transcribe audio from various sources:
  - Uploaded audio files
  - Direct URLs to MP3 files
  - YouTube video URLs
- Utilizes the latest GitHub version of Faster Whisper for optimal performance
- Adjustable batch size for performance tuning
- Provides detailed metrics including transcription time and real-time factor

## How to Use

1. Enter the source of your audio:
   - Path to a local audio file
   - URL of an MP3 file
   - URL of a YouTube video
2. Adjust the batch size if desired (default is 16)
3. Click 'Submit' to start the transcription process

## Output

The application will provide:
- A full transcription of the audio
- Detected language and confidence
- Duration of the audio
- Transcription time and real-time factor
- File size of the processed audio

## Note

This application is a prototype and may be subject to further improvements and optimizations. Performance may vary based on the input source and the processing capabilities of the hosting environment.

## Feedback and Contributions

I welcome feedback and contributions to improve this transcription tool.