Audio Transcription and Processing Tool

A Python application that transcribes audio files into text using OpenAI's Whisper model and processes transcripts using GPT-4o-mini. This tool supports various audio formats and allows for both audio transcription and transcript formatting.

Features

• Audio Transcription: Convert audio files into text using OpenAI's Whisper API.
• Transcript Processing: Enhance readability and structure of transcripts using GPT-4o-mini.
• Supports Multiple Audio Formats: Works with common audio file types.
• Modular Design: Clean and maintainable codebase with separated concerns.
• Output Management: Prevents overwriting by generating unique filenames and organizes outputs into dedicated directories.

Requirements

• Python 3.7 or higher
• An OpenAI API key with access to the Whisper and GPT-4o-mini models

Python Packages

Listed in requirements.txt:

• pydub
• openai
• tiktoken

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/yourrepository.git
   cd yourrepository
   

2. Create a virtual environment (optional but recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`
   

3. Install the required packages:

   pip install -r requirements.txt
   

Setup

OpenAI API Key

The application requires an OpenAI API key to access the transcription and language models.

1. Obtain an API Key:

   • Sign up for an account at OpenAI.
   • Navigate to the API Keys section.
   • Create a new secret key.

2. Set the API Key as an Environment Variable:

   • On macOS/Linux:

     export OPENAI_API_KEY='your-api-key-here'
     

   • On Windows (Command Prompt):

     set OPENAI_API_KEY='your-api-key-here'
     

   • On Windows (PowerShell):

     $env:OPENAI_API_KEY='your-api-key-here'
     

   Replace 'your-api-key-here' with your actual API key.

Usage

The application can process both audio files and existing transcripts.

Processing an Audio File

Transcribe an audio file and process the transcript.

python main.py path_to_audio_file

Example:

python main.py audio_files/meeting_recording.mp3

Processing a Transcript File

Process an existing transcript to enhance readability and structure.

python main.py path_to_transcript_file --transcript

Example:

python main.py transcripts/raw_transcript.txt --transcript

Supported Audio Formats

The application supports the following audio file formats:

• mp3
• mp4
• mpeg
• mpga
• m4a
• wav
• webm

Output

• Raw Transcripts:
  • Saved in the raw_transcripts/ directory.
  • Filenames include the audio file's name.
  • If a file with the same name exists, a numbered suffix is added (e.g., meeting_recording.txt, meeting_recording_1.txt).
• Processed Transcripts:
  • Saved in the transcripts/ directory.
  • Filenames include the base name of the input file.
  • Numbered suffixes are added to prevent overwriting existing files.

OpenAI Models Used

• Transcription: whisper-1
  • Used to transcribe audio files into text.
• Transcript Processing: gpt-4o-mini
  • Enhances the readability and structure of transcripts.

Application Structure

The application is designed with modularity in mind, separating concerns across different modules.

Modules

• main.py:
  • Entry point of the application.
  • Handles argument parsing and initializes the application.
• app.py:
  • Contains the TranscriberApp class.
  • Manages the core workflow for processing audio and transcript files.
• file_manager.py:
  • Handles file and directory operations.
  • Creates necessary directories and manages unique filenames.
• logger_config.py:
  • Configures logging for the application.
  • Ensures logs are output to the console.
• api_validator.py:
  • Validates the OpenAI API key by making a test API call.
• audio_splitter.py:
  • Splits large audio files into smaller chunks if necessary.
• audio_transcriber.py:
  • Transcribes audio files using the OpenAI Whisper API.
• transcript_processor.py:
  • Processes transcripts with GPT-4o-mini to enhance readability and structure.

Directory Structure

├── audio_files/           # Place your audio files here
├── raw_transcripts/       # Stores raw transcripts generated from audio files
├── transcripts/           # Stores processed transcripts
├── tmp/                   # Temporary files
├── main.py                # Entry point of the application
├── app.py                 # Core application logic
├── file_manager.py        # File and directory management
├── logger_config.py       # Logging configuration
├── api_validator.py       # API key validation
├── audio_splitter.py      # Audio splitting logic
├── audio_transcriber.py   # Audio transcription logic
├── transcript_processor.py# Transcript processing logic
├── requirements.txt       # Python package requirements