Voice Generator

A web application for generating audio from vocabulary lists using OpenAI's Text-to-Speech API.

Features

• Extract vocabulary from PDF files using OCR
• Create and manage vocabulary lists
• Generate audio files from vocabulary lists
• Support for multiple languages (Dutch, Turkish, Russian)
• Manual text input for vocabulary extraction
• Parallel processing for faster extraction (OCR and AI processing)
• Configurable parallel processing settings
• Rate limiting for API calls

Setup

1. Get an OpenAI API Key

   • Visit OpenAI API
   • Create a new API key
   • Make sure you have credits in your OpenAI account

2. Run the Application

   • Open index.html in your web browser
   • When prompted, enter your OpenAI API key
   • The API key will be stored in memory for the current session

Usage

Creating Vocabulary Lists

1. From PDF Files

   • Click "Upload PDF" and select a PDF file
   • Choose whether to include Russian translations
   • Enable OCR if the PDF contains images or scanned text
   • Configure parallel processing settings (optional):
     • Enable/disable parallel processing
     • Set max OCR workers (1-8)
     • Set max AI workers (1-16)
     • Adjust text chunk size (1000-5000 characters)
   • Click "Extract Vocabulary"

2. Manual Input

   • Click "Manual Input"
   • Paste or type your text
   • Click "Process Text"

3. Manual Creation

   • Click "Create New List"
   • Add vocabulary items manually
   • Save your list

Generating Audio

1. Select a vocabulary list
2. Choose your preferred voices for each language
3. Set the number of items per audio file
4. Choose generation mode (parallel or sequential)
5. Click "Generate Audio"
6. Download the generated audio files

Security

• API keys are never stored on disk
• API keys are only stored in browser memory during the session
• No sensitive data is sent to any server except OpenAI's API

Requirements

• Modern web browser with JavaScript enabled
• OpenAI API key with credits
• Internet connection for API calls

Troubleshooting

• API Key Error: Make sure your API key is valid and you have credits
• PDF Extraction Issues: Try enabling OCR for scanned documents
• Audio Generation Fails: Check your internet connection and API key validity
• Parallel Processing Issues:
  • Reduce the number of concurrent workers if you experience timeouts
  • For large PDFs, increase chunk size to reduce the number of API calls
  • Monitor your OpenAI API rate limits

File Structure

VoiceGenerator/
├── index.html          # Main application page
├── script.js           # Application logic
├── styles.css          # Styling
├── server.py           # Optional local server
├── package.json        # Dependencies (if using npm)
└── README.md           # This file