HearAI is a general-purpose local inference backend designed to run AI models entirely on your own machine.
It enables speech-to-text, meeting summarization, and local LLM-powered applications without sending any data to the cloud.
- 🎙️ Speech-to-Text using Whisper
- 🧠 Local LLMs (LLaMA, Mistral, etc.)
- 📄 Meeting Summaries & Action Items
- 🔒 100% Private — Runs Fully Offline
- ⚡ FastAPI-based High-Performance API
- 📱 Works with Mobile, Web & Desktop Apps
Mobile / Web / Desktop Apps ↓ HearAI API ↓ Whisper / LLM / AI Models ↓ Transcripts, Summaries, Actions
HearAI acts as the AI brain for all your local applications.
- FastAPI – REST API
- Whisper – Speech Recognition
- Local LLMs – LLaMA, Mistral, etc.
- MySQL – Memory, transcripts, sessions
- Python 3.12
- Conda – Environment management
conda create --name HearAI python=3.12 -y
conda activate HearAI2️⃣ Install Dependencies
pip install -r requirement.txt3️⃣ Setup Database
Login to MySQL:
mysql -u root -p
Create database:
CREATE DATABASE voice_ai;4️⃣ Configure Database
Open database.py and update:
mysql+pymysql://jay:StrongPassword123@localhost:3306/voice_ai
(Replace username and password if needed.)For local use:
uvicorn app.main:app --reloadFor access from mobile or other devices on your Wi-Fi:
uvicorn app.main:app --host 0.0.0.0 --port 8000The API will be available at:
http://localhost:8000or
http://<your-local-ip>:8000📡 What HearAI Is For
HearAI is not a single-purpose API. It is a general AI inference engine that can power:
Voice assistants
Meeting & call summarizers
Offline ChatGPT-like apps
Smart note-taking apps
Enterprise AI tools
🔒 Why HearAI? Feature Benefit Local execution No cloud, no data leaks Offline capable Works without internet Multi-model support Speech + LLM + NLP One backend Multiple client apps FastAPI Easy integration
🧠 Vision
HearAI is built to be the core AI engine for private, offline, and secure intelligent applications.
Your voice. Your data. Your AI.
If you want, next I can help you add:
• API endpoint docs
• Screenshots section
• Logo & branding
• Deployment guide
Just say 😎