VoiceForge AI Frontend

Modern, ultra-fast UI rendering interface built exclusively via React 19, Vite 7, and Tailwind CSS 4.

What This App Does

• Operates as the user-facing command layer rendering local voice cloning configurations directly.
• Handles logic bindings allowing fluid control of operations requesting generation to http://localhost:3000/api/clone-voice and http://localhost:3000/api/generate-video.
• Orchestrates multi-state inputs permitting flexible variations for the end user:
  • Voice Engine Selection: Qwen3-TTS, XTTSv2, or ElevenLabs API.
  • Input Method Selection: Live user microphone uploads vs. built-in standard AI actor IDs.
  • Render Mode Options: Generate Audio-Only outputs vs. full Portrait-Video renders.
  • Video Engine Configurations: Optimized Wav2Lip mode or High-Fidelity SadTalker mode.

Core Dependencies

• react, react-dom
• tailwindcss, @tailwindcss/vite
• vite
• Custom eslint enforcing logic

Prerequisites

• Node.js 20+
• A running orchestrator backend processing requests on port 3000.

Install

npm install

Run In Development

npm run dev

Default Application Location: http://localhost:5173

Production Build

npm run build

Repo Notes

• This zone tracks frontend configuration semantics exclusively.
• All artifact manifestations (dist/, node_modules/) are .gitignored outright avoiding footprint saturation.
• Pure display tier application logic strictly decoupled from machine learning models.