README.md

metadata

title: OMANI-Chatbot
emoji: 💻
colorFrom: blue
colorTo: blue
sdk: streamlit
sdk_version: 1.50.0
app_file: OMANI-Chatbot.py
pinned: false
license: mit

README.md

🇴🇲 OMANI-Chatbot

OMANI-Chatbot is a mental health–focused conversational assistant designed to provide supportive dialogue, intent recognition, and safe responses.
It integrates speech-to-text (STT), text-to-speech (TTS), retrieval-augmented generation (RAG), and emotion detection into a single pipeline.

---

📂 Project Structure

OMANI-Chatbot.py         # Main entrypoint
app/                     # Core application logic
data/                    # Knowledge base + Chroma vector database
documentation/           # Architecture diagrams and reports
tests/                   # Unit tests, audio samples, transcripts
requirements.txt         # Python dependencies

---

⚙️ Setup & Installation

1. Clone the repository

git clone https://github.com/russelljeffrey/Omani-therapist-voice.git

2. Create virtual environment & install dependencies

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

---

🚀 Running the Project

Option A: Run Locally

streamlit run OMANI-Chatbot.py

By default, the Streamlit UI will launch on http://localhost:8501

---

Option B: Run with Docker

1. Build the Docker image:

docker build -t omani-chatbot .

2. Run the container:

docker run -p 8501:8501 omani-chatbot

Open your browser at http://localhost:8501 to use the chatbot.

---

📖 Documentation

See the documentation/ folder for:

• architecture.md → System overview
• safety_protocol.md → Safety design
• future_roadmap.md → Planned improvements
• wer_report.md → Performance metrics

---

🧪 Testing

Run the tests with:

python -m pytest -s -v tests/

---

🛡️ Safety Note

This project is for demo purposes. It is not a substitute for professional mental health care. If someone is in crisis, they should seek immediate professional support.

---