MuhirwaVerygood/RwandaVoiceAI-
Domain:
natural language processingRecord type:
softwareproject
Creator:
Muh
Host:
🎙️ Ask questions in Kinyarwanda using your voice and get instant AI-powered responses — spoken back to you and displayed as text! 🇷🇼 Powered by speech recognition, natural language understanding, and text-to-speech. Speak. Hear. See.
# 🤖 Kinyarwanda Voice Assistant
An intelligent voice assistant that understands and speaks **Kinyarwanda**, developed as part of the **Intelligent Robotics** course. It combines advanced speech recognition, contextual understanding, and natural language response generation to enable real-time human-computer interaction in Kinyarwanda.
---
## 🎥 Interface Demo
A web-based interactive demo is included using **Gradio**. Launch the app locally and interact with it through your browser.
---
## 🌟 Features
- 🎙️ **Kinyarwanda ASR** powered by `KinyaWhisper` (optimized for 16kHz audio)
- 🧠 **Contextual Understanding** using fuzzy logic matching
- 📢 **Natural Language Responses** with Kinyarwanda Text-to-Speech (TTS)
- 🔇 **Noise Reduction** with advanced audio cleaning
- 🎚️ **Voice Activity Detection (VAD)** for precise speech segmentation
- 🔄 **Repetition Filters** for clean, intelligible transcription
- 📊 **Conversation Analytics** with question matching insights
- 🌐 **Web Interface** via Gradio for user-friendly interaction
---
## 🛠️ Tech Stack
| 🧩 Layer | 🛠️ Tools Used |
|---------------------|--------------------------------------|
| Core AI | Hugging Face Transformers |
| Audio Processing | Librosa, Soundfile, WebRTC VAD |
| NLP Matching | FuzzyWuzzy, Python-Levenshtein |
| Voice Interface | Gradio |
| Optimization | Noisereduce, FFmpeg |
---
## 💻 Installation
### ✅ Prerequisites
- Python 3.12
- FFmpeg
Install FFmpeg:
```bash
# Ubuntu/Debian
sudo apt-get install ffmpeg
# macOS
brew install ffmpeg
# Windows (via Chocolatey)
choco install ffmpeg
```
---
### 🚀 Quick Start
```bash
# Clone the repository
git clone github.com
cd RwandaVoiceAI--main/RwandaVoiceAI--main
# Set up virtual environment
python -m venv .venv
# Activate virtual environment
.\.venv\Scripts\activate …