Àkàndé

Àkàndé is an advanced voice assistant built in Python, leveraging OpenAI's GPT models for natural language understanding and response generation. Àkàndé has been enhanced to include a caching mechanism for efficient response retrieval and the ability to generate PDF summaries of interactions, making it ideal for both personal assistance and executive briefing purposes.

Features

• Natural Language Understanding: Utilizes OpenAI's GPT models to understand and generate human-like responses.
• PDF Summary Generation: Generates PDF summaries of voice interactions, including a question header, AI-generated response, and an accompanying logo.
• Caching Mechanism: Implements a SQLite-based caching system to store and retrieve past queries and responses, reducing API calls and improving response times.
• Voice Recognition: Integrates with speech recognition libraries to support voice input.
• Text-to-Speech: Converts text responses into speech, providing an interactive voice-based user experience.

Setup

Prerequisites

• Python 3.8+
• Pipenv or virtualenv

Installation

1. Clone the repository

git clone https://github.com/sebastienrousseau/akande
cd akande

2. Install dependencies

pipenv install  # If using pipenv
# or
python -m venv venv
source venv/bin/activate  # On Windows use `venv\Scripts\activate`
pip install -r requirements.txt

3. Set up environment variables

Copy .env.example to .env and fill in your OpenAI API key and other configurations.

OPENAI_API_KEY=xxxxxxxxxx

4. Running Àkàndé

pipenv run python -m akande  # If using pipenv
# or
python -m akande

Usage

After starting Àkàndé, simply follow the voice prompts to ask questions. Àkàndé will respond verbally and generate a PDF summary for each interaction in the specified output directory.

Contributing

Pull requests are welcome. See CONTRIBUTING.md for guidelines.

License

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