Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

VoxCode — Professional Voice-To-Code Assistant

VoxCode is a production-grade, advanced-level voice-controlled programming assistant. It enables developers to generate, optimize, and explain code using natural voice commands, bridging the gap between human thought and executable code with a high-performance, dual-engine speech pipeline.

🚀 Advanced Architecture

🧠 Backend (Python / Flask)

  • API Server: Hardened Flask REST API with structured logging, per-IP rate limiting, and DB connection pooling for high concurrency.
  • Speech Pipeline: Hybrid engine using Vosk (real-time partials) and Whisper (high-accuracy final transcription). Includes WebRTC VAD for voice activity detection and Noisereduce for adaptive environmental noise profiling.
  • Command Processor: Advanced intent detection using a hybrid of regex pattern matching and Zero-Shot BART Transformers. Supports multi-intent detection and contextual memory for follow-up commands (e.g., "convert that to rust").
  • NLP Service: Deep linguistic analysis with spaCy. Features a coding-specific dictionary (100+ entries) and abbreviation expansion (e.g., fnfunction).
  • Template Service: Idiomatic code scaffolding for 12+ languages including Rust, Go, Swift, and Kotlin.

🎨 Frontend (Next.js 16 App Router)

  • Framework: Next.js 16 (App Router) with React 19, Server-Side Rendering (SSR), React <Suspense> boundaries, and dynamic code splitting.
  • Theme System: Custom "Light & Tiny Dark" aesthetic featuring crisp light canvas backgrounds (#FAFBFC) paired with refined, sleek dark charcoal/ink accents across sidebars, IDE consoles, and terminal panels.
  • Dashboard & IDE: Pro Studio with Monaco Editor, interactive 4-mode execution compiler (AST audit, RAM trace, Unit tests), and real-time audio waveform visualizer.
  • Interactions: Toast-based notification system, step-by-step code explanation mode, and 45+ domain-specific voice shortcuts.

🛠️ Stack

  • Backend: Python 3.11, Flask, PostgreSQL (NeonDB), Whisper, Vosk, spaCy, BART, NLTK.
  • Frontend: Next.js 16 (App Router), React 19, Tailwind CSS v4, Lucide icons, Monaco Editor, Framer Motion.
  • DevOps: Local LLM hosting (Llama-CPP), structured logging, health metrics.

📦 Installation

  1. Dependencies:
    .\install_dependencies.bat
  2. Configuration: Copy .env.example to .env and set your DATABASE_URL and JWT_SECRET.
  3. Start:
    .\start.bat

🎤 Core Voice Shortcuts

  • "Create a linked list in rust"
  • "Quick sort algorithm in go"
  • "Build a flask api with jwt auth"
  • "React component named UserCard"
  • "Optimize this code"
  • "Explain how this function works"

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages