Airline Assistant Demo App

A conversational AI assistant that helps users manage their airline reservations through natural language and voice interactions. This demo application supports multiple speech recognition and text-to-speech providers for a comprehensive voice-enabled experience.

Features

✈️ Flight Management

• Find flight information using natural language queries
• Look up airline ticket policies
• Update and modify flight bookings
• Check flight schedules and availability

🎤 Voice Interaction

• Voice-to-text input for hands-free operation
• Text-to-speech responses for accessibility
• Support for multiple languages

🔧 Multiple AI Providers

• Speech Recognition (ASR): Google, Azure, Fanolab
• Text-to-Speech (TTS): Google, Azure, Fanolab
• Easy switching between providers in the application

Quick Start

1. Installation

Clone the repository and install dependencies:

git clone https://github.com/jackychh7878/FlightAssistantAgent.git
cd FlightAssistantAgent
pip install -r requirements.txt

2. Configuration

Before running the app, ensure you have API keys configured for your preferred ASR/TTS providers:

• Google Cloud: Set up Google Cloud credentials for Speech-to-Text and Text-to-Speech APIs
• Azure Cognitive Services: Configure Azure Speech Services API key
• Fanolab: Set up Fanolab API credentials

⚠️ Important: Make sure you have:

• Valid API keys for your chosen providers
• Sufficient credits/quota for your selected services
• Proper authentication configured

3. Run the Application

Start the Streamlit app:

streamlit run app.py

The application will be available at http://localhost:8501

Usage

Demo Reset

🔄 Note: Each time the application starts, it automatically resets the SQLite database to ensure a clean demo environment for new users.

Voice Mode Configuration

In the app interface, you can switch between different ASR and TTS providers:

1. Google Cloud: High-quality speech recognition and synthesis
2. Azure Cognitive Services: Microsoft's speech services
3. Fanolab: Specialized speech processing services

Make sure your selected provider has:

• ✅ Valid API credentials configured
• ✅ Sufficient account credits
• ✅ Proper network connectivity

Sample Questions

Here are some example queries you can try (in Cantonese):

我最新的航班係幾時?
(When is my latest flight?)

如果我想轉去下星期, 最快既期有邊一日?
(If I want to change to next week, what's the earliest available date?)

幫我改期
(Help me change the date)

如果改機票要幾多錢?
(How much does it cost to change the flight ticket?)

Technical Architecture

Core Components

• Frontend: Streamlit web interface
• Backend: Python-based conversational AI
• Database: SQLite for flight data storage
• AI Integration: LangChain for natural language processing
• Voice Processing: Multiple ASR/TTS provider integrations

Supported File Structure

AKAAgenticAI/
├── app.py                      # Main Streamlit application
├── requirements.txt            # Python dependencies
├── src/
│   ├── azure_asr.py           # Azure Speech Recognition
│   ├── azure_tts.py           # Azure Text-to-Speech
│   ├── google_stt.py          # Google Speech-to-Text
│   ├── google_tts.py          # Google Text-to-Speech
│   ├── fanolab_asr_tts.py     # Fanolab ASR/TTS
│   ├── sqlite_tools.py        # Database utilities
│   └── vector_store_retriever.py # RAG implementation
└── research/                   # Development notebooks

Troubleshooting

Common Issues

1. API Key Errors

   • Verify your API keys are correctly configured
   • Check if you have sufficient credits/quota
   • Ensure proper authentication setup

2. Audio Issues

   • Check microphone permissions in your browser
   • Verify audio input/output devices are working
   • Test with different browsers if needed

3. Database Issues

   • The SQLite database resets on each startup (by design)
   • Check file permissions in the src/sqlite_db/ directory

Provider-Specific Setup

Google Cloud

1. Create a Google Cloud project
2. Enable Speech-to-Text and Text-to-Speech APIs
3. Create service account credentials
4. Set GOOGLE_APPLICATION_CREDENTIALS environment variable

Azure Cognitive Services

1. Create Azure Cognitive Services resource
2. Get your subscription key and region
3. Configure environment variables for Azure Speech SDK

Fanolab

1. Register for Fanolab services
2. Obtain API credentials
3. Configure authentication in the application

Development

For development and customization:

1. Adding New Providers: Extend the ASR/TTS modules in the src/ directory
2. Database Schema: Modify sqlite_setup.py for data structure changes
3. UI Customization: Update app.py for interface modifications
4. Voice Processing: Enhance audio handling in respective provider modules

License

[Add your license information here]

Support

For questions or issues, please [add contact information or issue tracker link].