🛡️ Privacy-First • 🚀 Lightning-Fast • 🤖 AI-Powered • 📱 Cross-Platform
Complete offline transcription with speaker identification, AI summaries, and export capabilities
|
🛡️ 100% Privacy-First - All AI processing happens offline |
🎓 Students - Record lectures, meetings, interviews |
| Backend Powerhouse | Frontend Excellence | AI & Processing |
|---|---|---|
 |
 |
 |
 |
 |
 |
 |
 |
 |
# 🔥 Step 1: Clone the repo
git clone https://github.com/OP-88/Verba.devops.git
cd Verba.devops
# 🚀 Step 2: Backend setup
cd backend
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install -r requirements.txt
# ⚡ Step 3: Start backend server
uvicorn src.run_fastapi_audio_fixed:app --reload --host 0.0.0.0 --port 8000
# 🎨 Step 4: Frontend setup (new terminal)
cd frontend
export VITE_API_URL=http://localhost:8000 # Windows: set VITE_API_URL=http://localhost:8000
npm install && npm run dev🎉 Visit http://localhost:8080 and start transcribing! 🎉
🌐 Try the Live Demo • 📝 Read the Docs • 🐛 Report Issues
- ⚡ Real-Time Transcription - Live microphone recording with instant text conversion
- 📁 File Upload Support - Process WAV, MP3, M4A, and more audio formats
- 🔊 Smart VAD - Enhanced voice activity detection with Silero VAD
- 🎯 Noise Reduction - Advanced audio preprocessing for clarity
- 📊 Audio Visualization - Real-time waveform and level monitoring
- 🎙️ Speaker Diarization - Automatic "who said what" identification using pyannote.audio
- 📊 Speaker Statistics - Speaking time analysis and dominant speaker detection
- 🏷️ Smart Labeling - Automatic speaker assignment to transcript segments
- 🔄 Segment Merging - Intelligent combining of short speech segments
- 📝 Auto-Summarization - T5-powered summaries with key points extraction
- 🎯 Action Items - Automatic detection of tasks and follow-ups
- 📈 Sentiment Analysis - Meeting tone and mood detection
- 💬 Smart Chat - AI assistant for transcript queries (hybrid mode)
- 📄 Multiple Formats - Markdown, PDF, JSON, TXT, SRT with full metadata
- ⚙️ Customizable Exports - Include/exclude metadata, speakers, summaries
- 📋 One-Click Copy - Instant clipboard access with formatting
- 💾 Auto-Save - SQLite database with full history tracking
- Ctrl+R - Start/Stop recording
- Ctrl+P - Pause/Resume recording
- Ctrl+C - Copy transcription
- Ctrl+E - Edit transcription
- Ctrl+S - Save/Export
- Esc - Cancel current action
- 🌐 Web App - Modern React interface with PWA support
- 🖥️ Desktop Apps - Native Tauri applications for Windows, macOS, Linux
- 📱 Mobile Responsive - Touch-optimized interface for tablets and phones
- ☁️ Cloud Deploy - One-click Vercel deployment ready
graph TB
A[🎤 Audio Input] --> B[🌊 WebRTC Stream]
B --> C[⚡ FastAPI Backend]
C --> D[🤖 Whisper AI]
D --> E[📝 Transcription]
E --> F[💾 SQLite Storage]
F --> G[📱 React Frontend]
G --> H[👤 Beautiful UI]
🏠 verba/
├── 🚀 backend/ # FastAPI powerhouse
│ ├── 🎯 main.py # Server magic starts here
│ ├── 🗃️ models/ # Database schemas
│ ├── 🛣️ routes/ # API endpoints
│ ├── ⚙️ services/ # Whisper AI integration
│ └── 📋 requirements.txt
├── 💎 frontend/ # React brilliance
│ ├── 🎨 src/
│ │ ├── 🧩 components/ # Reusable UI magic
│ │ ├── 📄 pages/ # Main app screens
│ │ ├── 🔗 services/ # API communication
│ │ └── 🎯 types/ # TypeScript definitions
│ ├── 📦 package.json
│ └── ⚡ vite.config.ts
└── 📚 docs/ # Everything you need to know
| 🚀 Method | 🎯 Endpoint | 💡 What It Does | ✨ Magic |
|---|---|---|---|
GET |
/health |
💚 Server heartbeat | Always alive |
POST |
/transcribe |
🎤 Transform audio → text | AI-powered |
GET |
/history |
📜 Your transcription story | Full history |
POST |
/history |
💾 Save your gems | Instant storage |
DELETE |
/history/{id} |
🗑️ Clean up | One-click delete |
GET |
/export/{id} |
📤 Download magic | Multiple formats |
| 🎯 Endpoint | 💫 Real-Time Magic |
|---|---|
/ws/transcribe |
⚡ Live transcription stream |
|
|
💾 RAM: 4GB
💿 Storage: 2GB free
⚡ CPU: Dual-core
🌐 Browser: Chrome 80+ | Firefox 75+ | Safari 13+ |
🔥 RAM: 8GB+
💿 Storage: 5GB free
⚡ CPU: Quad-core+
🎮 GPU: CUDA-compatible (optional boost!) |
gantt
title 🚀 Verba Development Timeline
dateFormat YYYY-MM-DD
section 🏗️ Foundation
Backend API Core :active, 2024-09-15, 7d
Database Schema :active, 2024-09-16, 5d
Whisper Integration :2024-09-20, 4d
section 🎨 Frontend
React UI Base :2024-09-18, 6d
WebRTC Recording :2024-09-22, 5d
Real-time Display :2024-09-25, 4d
section ✨ Polish
Export Features :2024-09-28, 3d
UI/UX Enhancement :2024-09-30, 5d
Testing & Deploy :2024-10-03, 4d
| Phase | Feature | Status | Timeline |
|---|---|---|---|
| 🏗️ | Core API | 🔄 In Progress | Week 1 |
| 🤖 | Whisper AI | ⏳ Planned | Week 2 |
| 🎨 | React UI | 🔄 In Progress | Week 2 |
| ⚡ | Real-time | ⏳ Planned | Week 3 |
| 💎 | Export | ⏳ Planned | Week 4 |
# 🍴 Fork it
git clone https://github.com/YOUR-USERNAME/Verba.devops.git
# 🌱 Branch it
git checkout -b feature/amazing-transcription-magic
# ✨ Code it
# ... your brilliant contributions ...
# 🚀 Push it
git push origin feature/amazing-transcription-magic
# 🎉 PR it - Open a Pull Request!|
🎨 Frontend Magic
|
⚡ Backend Power
|
🤖 AI Enhancement
|
Coming soon - your name could be here!
Be the first to contribute and earn your place in Verba history! 🚀
| 🐛 Issue | 💡 Status | 🎯 Solution |
|---|---|---|
| Repository URL verification | 🔄 Working | Testing clone process |
| Development environment | ⚡ Priority | Automated setup script |
| Dependency management | 🔄 Active | Version compatibility check |
|
This project is licensed under the MIT License - see the LICENSE file for details.
🎉 Free to use, modify, and distribute! 🎉
Transforming the way we interact with audio, one transcription at a time
🚀 Ready to revolutionize transcription? Let's build something amazing together! 🚀
Made with 🔥 passion and ⚡ cutting-edge technology