voxai 0.3.3

Voice-driven AI assistant with modern UI, system audio capture, and ChatGPT-like streaming responses.

VoxAI

🚀 Features

• 🎧 System Audio Capture
  Captures clean audio from your computer (meetings, browser, apps) - no external microphone noise or room interference.
• 🎮 Manual Control
  Start/Stop buttons let you define exactly the boundaries of your prompt—perfect for long, multi-sentence queries.
• 🔄 Optimized Transcription
  Fast Whisper tiny model with GPU acceleration for near real-time speech-to-text processing.
• ⚡ Live AI Streaming
  ChatGPT-like streaming responses with structured formatting, headings, and bullet points.
• 🎯 Concise Technical Responses
  Structured responses with bullet points optimized for quick technical understanding.
• 🔧 Configurable Model
  Swap between Gemini variants via environment (no code changes).
• 🚀 Zero-Install UI Bootstrapping
  On first run voxai auto-installs Electron dependencies—included in the PyPI package—so you only ever need pip install voxai and voxai.
• 🎨 Modern UI
  Beautiful gradient interface with real-time status indicators, typing animations, and professional styling.

---

🎯 Quickstart

⚙️ Prerequisites

• Python ≥3.7
• Node.js ≥14 (for Electron UI)

That's it! VoxAI automatically installs Electron dependencies on first run.

Important: VoxAI captures system audio only (meetings, browser, apps) - no external microphone.

Install & Run

1) Install from PyPI

pip install voxai

2) Configure environment

cp .env.example .env

Edit .env:

GENAI_API_KEY=sk-…

GENAI_MODEL=gemini-1.5-flash

3) Launch the app

voxai

On first launch, VoxAI will automatically run npm install inside its bundled electron/ folder, then open a desktop window.

---

📖 Usage

Once the UI opens:

Start: Click Start to begin recording.

Speak: Listen to question`s from zoom, teams or any other tool aloud—no length limit.

Stop: Click Stop. Whisper transcribes your entire clip to the Transcript pane.

Ask AI: Click Ask AI. Gemini’s answer streams live into the Answer pane.

Copy & Share: Everything is plain text—copy it, paste it, or feed it into your own RAG/finetuning pipeline.

---

🎧 How It Works

• 🎤 Audio Source Detection
  VoxAI automatically detects your system audio setup and displays it in the beautiful UI with real-time status.

• 🎬 Start Recording
  Click Start Recording to capture clean audio from your computer (meetings, browser, apps).

• ✋ Stop & Transcribe
  Click Stop Recording. Whisper transcribes the entire audio clip and shows it under Transcript.

• 🤖 Ask AI
  Click Ask AI to send the transcript to Gemini. Get ChatGPT-like streaming responses with structured formatting.

• 📋 Copy & Share
  Professional responses with headings, bullet points, and technical depth—perfect for documentation or sharing.

🎨 Modern UI Features

• 🌈 Beautiful Design: Gradient backgrounds with glassmorphism effects
• 💫 Real-time Animations: Typing indicators, loading states, and smooth transitions
• 📊 Status Indicators: Color-coded status dots (🟢 ready, 🔴 recording, 🟡 thinking)
• 📱 Responsive Layout: Works perfectly on different screen sizes
• 🎯 Professional Formatting: AI responses with headings, bullets, and technical structure

---

🛠 Configuration

Set in .env:

Variable	Description	Example
GENAI_API_KEY	Google Generative AI (Gemini) API key	sk-…
GENAI_MODEL	Gemini model to use	gemini-1.5-flash

See .env.example for reference.

---

🛠️ Developer Guide

Install from source

git clone https://github.com/rtiwariops/voxai.git

cd voxai

pip install -e .

npm install -g electron

voxai

🔧 System Audio Setup

VoxAI captures computer audio only (no microphone) - perfect for meeting recordings, browser audio, and app sounds without room noise.

macOS Setup

# 1. Install BlackHole
brew install blackhole-2ch

# 2. Install Ladiocast
# Download from: https://existential.audio/ladiocast/

Configure Ladiocast:

1. Input 1: Set to your audio source (Built-in Input or system audio)
2. Main: Route Input 1 to Main (for speakers)
3. Aux 1: Set to BlackHole 2ch
4. Enable: Route Input 1 to Aux 1

System Settings:

• Input: BlackHole 2ch (VoxAI reads from here)
• Output: Built-in Output (you hear from here)

Windows (Built-in)

WASAPI loopback support is built into Windows 10+ - VoxAI will auto-detect.

Linux (Built-in)

PulseAudio monitor devices are auto-detected - no setup needed.

Result: You hear audio through speakers + VoxAI captures clean computer audio.

---

⚙️ Features

• Smart Cross-Platform Audio Detection
• Manual Control for long-form Q&A
• One-Shot, Full-Clip Transcription
• Live Token-by-Token AI Streaming
• Configurable Gemini Model (.env)
• Electron-Based Desktop UI

---

📜 License

MIT License © 2025 Ravi (Robbie) Tiwari

Released under the MIT License.