PAL - Personal AI Launcher

PAL is your desktop co-pilot: fast voice chat, sharp text reasoning, and a clean Tauri-native experience that stays close to your workflow instead of living in a browser tab.

---

Tech Used 🧑‍💻

---

Core Features ⚡

• 🎙️ Voice-First Assistant: Talk naturally with PAL using live voice input and spoken output.

• 💬 Text + Voice Hybrid Chat: Switch between typing and speaking without breaking the conversation flow.

• 🧠 Multiple Assistant Modes: Different personalities/work modes for different tasks.

• 🕘 History + Stats: Track past chats and usage analytics from inside the app.

• ⌨️ Global Shortcuts + Tray Control: Summon PAL quickly, then hide/show/quit from the system tray.

• 💾 Persistent Local Settings: Startup behavior, preferences, and app state are saved locally.

• 🖥️ Desktop-Native Performance: Tauri + Rust backend for a lightweight, responsive experience.

---

Screenshots 📸

Voice Screen: Push-to-talk and conversational voice pipeline for fast hands-free interaction.

Chat Screen: Structured responses, markdown support, and focused conversation layout.

History Screen: Jump back into previous sessions instantly.

Stats Screen: Lightweight analytics to understand how you use PAL.

Settings Screen: Configure startup, models, voice behavior, and app preferences.

About Screen: Quick project overview and build context.

---

Runtime Modes 🧪

Current default build runs with cloud providers:

Flag	Value	Meaning
LOCAL_LLM	false	Chat model runs via cloud API
STT_LOCAL	false	Speech-to-text runs via cloud API
TTS_LOCAL	false	Text-to-speech runs via cloud API

Default model setup:

• Chat: llama-3.3-70b-versatile
• STT: whisper-large-v3-turbo
• TTS: canopylabs/orpheus-v1-english

---

Project Structure

/ (root)
├── README.md
├── AnotherProject.md
├── package.json
├── vite.config.ts
├── screenshots/            # App screenshots used in this README
├── public/                 # Static assets (including PAL icon)
├── src/                    # React + TypeScript frontend
│   ├── components/
│   ├── routes/
│   ├── services/
│   ├── styles/
│   ├── App.tsx
│   └── main.tsx
├── src-tauri/              # Rust + Tauri desktop backend
└── backend/                # Optional/local experimentation backend

---

Setup and Development 🛠️

1. Prerequisites:

   • Node.js (v18+)
   • Rust toolchain
   • Tauri CLI

2. Install dependencies:

   pnpm install

3. Configure environment: Create/update src/.env:

   LOCAL_LLM=false
   TTS_LOCAL=false
   STT_LOCAL=false
   
   VITE_GROQ_API_KEY=your_groq_key
   VITE_GROQ_BASE_URL=https://api.groq.com/openai/v1
   VITE_GROQ_CHAT_MODEL=llama-3.3-70b-versatile
   VITE_GROQ_STT_MODEL=whisper-large-v3-turbo
   VITE_GROQ_TTS_MODEL=canopylabs/orpheus-v1-english
   VITE_GROQ_TTS_VOICE=troy

4. Run in development:

   pnpm tauri dev

5. Other useful commands:

   pnpm dev
   pnpm build

---

Recommended IDE Setup 💻

• VS Code
• Tauri for VS Code
• rust-analyzer

---

Roadmap 🗺️

Phase 1: Core Experience

• Voice + text assistant workflows
• Tauri desktop integration
• Tray behavior and global shortcuts
• Start with OS

Phase 2: Expansion

• More local model runtime options
• Deeper assistant mode customization
• Improved session intelligence and memory controls

---

Notes

• Current setup is Windows-first.
• Active flow does not require a Python backend.
• App data is persisted locally via Tauri plugins.

---

Contact 📬

• GitHub: mohaneddz
• Email: mohaned.manaa.dev@gmail.com