Offline AI Assistant

A comprehensive offline AI assistant that provides Text-to-Speech (TTS), Speech-to-Text (STT), sentiment analysis, and document analysis capabilities without requiring an internet connection.

🚀 Features

Core Capabilities

• Text-to-Speech (TTS): Convert text to speech using pyttsx3
• Speech-to-Text (STT): Convert speech to text using speech recognition
• Sentiment Analysis: Analyze text sentiment using VADER
• Document Analysis: Analyze documents using Gensim (topic modeling, word frequency, etc.)
• 100% Offline: All processing done locally without internet dependency

Operating Modes

• Voice Mode: Full voice interaction (TTS + STT)
• Text Mode: Text-based interaction for environments without audio

📋 Requirements

System Requirements

• Python 3.7 or higher
• Windows 10/11, macOS, or Linux
• Microphone and speakers (for voice mode)
• At least 2GB RAM
• 500MB free disk space

Python Dependencies

• pyttsx3 (TTS)
• SpeechRecognition (STT)
• PyAudio (Audio processing)
• vaderSentiment (Sentiment analysis)
• gensim (Document analysis)
• numpy, scipy (Numerical processing)

🛠️ Installation

1. Clone or Download

git clone <repository-url>
cd offline-ai-assistant

2. Install Python Dependencies

pip install -r requirements.txt

3. Platform-Specific Setup

Windows

• PyAudio should install automatically
• TTS uses Windows SAPI voices
• No additional setup required

macOS

# Install portaudio (required for PyAudio)
brew install portaudio

# Install dependencies
pip install -r requirements.txt

Linux (Ubuntu/Debian)

# Install system dependencies
sudo apt-get update
sudo apt-get install python3-pyaudio portaudio19-dev espeak-ng

# Install Python dependencies
pip install -r requirements.txt

Linux (CentOS/RHEL/Fedora)

# Install system dependencies
sudo yum install portaudio-devel espeak-ng
# or for Fedora: sudo dnf install portaudio-devel espeak-ng

# Install Python dependencies
pip install -r requirements.txt

🚀 Usage

Running the Assistant

python main.py

Mode Selection

The assistant will prompt you to choose between:

1. Voice Mode: Full voice interaction
2. Text Mode: Text-based interaction

Available Commands

Basic Commands

• hello / hi - Greeting
• time - Get current time
• date - Get current date
• help - Show available commands
• exit / quit - Close the assistant

Sentiment Analysis

• sentiment or analyze - Analyze text sentiment
• The assistant will prompt you to enter text for analysis

Document Analysis

• document or analyze document - Analyze document content
• The assistant will prompt you to enter text for analysis

🔧 Configuration

TTS Settings

You can modify TTS settings in the initialize_tts() method:

• Speech rate (default: 150)
• Volume level (default: 0.9)
• Voice selection

STT Settings

Adjust STT parameters in the initialize_stt() method:

• Energy threshold (default: 4000)
• Dynamic energy threshold (default: True)
• Pause threshold (default: 0.8)

Audio Settings

Modify audio parameters in the __init__ method:

• Sample rate (default: 44100 Hz)
• Channels (default: 1)
• Chunk size (default: 1024)

📊 Features in Detail

Text-to-Speech (TTS)

• Engine: pyttsx3 with system TTS backends
• Voices: Automatically selects best available voice
• Customization: Adjustable speed and volume
• Languages: Supports multiple languages based on system voices

Speech-to-Text (STT)

• Engine: SpeechRecognition with Google's speech recognition
• Noise Reduction: Automatic ambient noise adjustment
• Timeout Handling: Configurable listening timeouts
• Error Handling: Graceful fallback for recognition failures

Sentiment Analysis

• Engine: VADER (Valence Aware Dictionary and sEntiment Reasoner)
• Features:
  • Positive, negative, and neutral sentiment scores
  • Compound sentiment score (-1 to +1)
  • Sentiment categorization
• Use Cases: Social media analysis, customer feedback, content analysis

Document Analysis

• Engine: Gensim
• Features:
  • Topic modeling (LDA)
  • Word frequency analysis
  • Key topic extraction
  • Document statistics
• Use Cases: Content analysis, document clustering, topic discovery

🐛 Troubleshooting

Common Issues

TTS Not Working

• Check if system has TTS voices installed
• On Linux, install espeak-ng: sudo apt-get install espeak-ng
• Verify pyttsx3 installation: pip show pyttsx3

STT Not Working

• Check microphone permissions
• Verify PyAudio installation: pip show PyAudio
• On Linux, install portaudio: sudo apt-get install portaudio19-dev

Import Errors

• Ensure all dependencies are installed: pip install -r requirements.txt
• Check Python version: python --version
• Verify virtual environment if using one

Audio Quality Issues

• Adjust microphone settings in system preferences
• Modify audio parameters in the code
• Check audio drivers and hardware

Performance Optimization

• Reduce TTS speech rate for better clarity
• Adjust STT energy threshold for your environment
• Use text mode for faster processing without audio

🔒 Privacy & Security

• 100% Offline: No data sent to external services
• Local Processing: All analysis done on your device
• No Logging: Optional logging can be disabled
• Secure: No network communication required

📝 Logging

The assistant creates log files for debugging:

• File: ai_assistant.log
• Console: Real-time logging to terminal
• Level: INFO by default

To disable logging, modify the setup_logging() method.

🤝 Contributing

Contributions are welcome! Areas for improvement:

• Additional TTS engines
• More STT options
• Enhanced sentiment analysis
• Better document analysis algorithms
• UI improvements
• Performance optimizations

📄 License

This project is open source. Please check the license file for details.

🙏 Acknowledgments

• pyttsx3: Cross-platform TTS library
• SpeechRecognition: Speech recognition library
• VADER: Sentiment analysis tool
• Gensim: Topic modeling and document analysis
• PyAudio: Audio processing library

📞 Support

For issues and questions:

1. Check the troubleshooting section
2. Review the logs in ai_assistant.log
3. Verify all dependencies are installed
4. Check system audio settings

---

Note: This assistant is designed to work completely offline. For best results, ensure your system has good quality TTS voices and a working microphone.