🎙️ RelayInsights: AI Audio Intelligence & Analytics

RelayInsights is a powerful, AI-driven audio transcription and analysis platform. It transforms raw audio recordings or uploaded files into actionable intelligence using state-of-the-art models from OpenAI. Whether it's meeting minutes, sales coaching, or sentiment analysis, RelayInsights provides deeper clarity from your audio data.

---

🚀 Key Features

• Multi-Modal Input: Upload audio files (MP3, WAV, M4A) or record directly within the browser.
• High-Fidelity Transcription: Powered by OpenAI Whisper for industry-leading accuracy.
• Intelligent Templates: Choose from 15+ pre-configured AI templates including:
  • Executive Summaries
  • Action Item Extraction
  • Sentiment Analysis
  • Meeting Minutes
  • SWOT Analysis
  • Blog Post Conversion
• Interactive Workbench: View, edit, and download both transcripts and AI-generated insights.
• Secure & Private: Input your own OpenAI API key for direct control over your data and costs.

---

🛠️ Technical Stack

• Frontend: Streamlit
• Transcription: OpenAI Whisper
• Insight Engine: OpenAI GPT-4o
• Core Logic: Python 3.8+

---

🚦 Getting Started

Prerequisites

• Python 3.8 or higher
• An OpenAI API Key (Get one here)

Installation

1. Clone the repository:

   git clone https://github.com/alphatechlogics/RelayInsights.git
   cd RelayInsights

2. Create a virtual environment:

   python -m venv env
   .\env\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

Configuration

Create a .env file in the root directory (optional, or input directly in UI):

OPENAI_API_KEY=your_api_key_here

---

📖 User Guide

1. Launch the App:

   streamlit run main.py

2. Setup: Enter your OpenAI API Key in the sidebar.
3. Upload/Record: Drag and drop an audio file or use the built-in microphone.
4. Transcribe: The app will automatically generate a transcript.
5. Analyze: Select an Insight Template from the dropdown and click "Generate Insights".
6. Export: Use the download buttons to save your results as .txt or .md files.

---

👩‍💻 Developer Guide

Project Structure

RelayInsights/
├── src/
│   ├── audio_processor.py  # Whisper API integration
│   ├── llm_engine.py       # GPT analysis logic
│   └── utils.py          # File management & cleanup
├── config/
│   └── prompts.json        # AI prompt templates
├── assets/
│   └── styles.css          # Custom UI styling
├── main.py                 # Streamlit application entry point
└── requirements.txt        # Project dependencies

Core Modules

• AudioProcessor: Handles communication with the OpenAI Whisper API. It manages file reading and transcription requests.
• LLMEngine: Manages prompting and responses from GPT models. It injects transcripts into predefined templates from prompts.json.

Adding New Templates

To add a new analysis type, simply append a new JSON object to config/prompts.json:

{
    "id": "my_new_template",
    "name": "Custom Analysis",
    "template": "Analyze the following transcript for [X]:\n\nTranscript:\n{transcript}"
}

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

---

Built with ❤️ by Alphatech Works