Skip to content

README_EN.md

Latest commit

 

History

History
144 lines (109 loc) · 4.58 KB

README_EN.md

File metadata and controls

144 lines (109 loc) · 4.58 KB
Raw

Voke

Voke · AI Voice Transcription Assistant License Platform

中文 | English


Voke is a cross-platform application based on Electron, supporting Windows, macOS, and Linux systems. Combining ASR (Automatic Speech Recognition) with LLM (Large Language Models), it not only "transcribes" but also "understands", "polishes", and "answers", helping you efficiently produce high-quality text.

✨ Core Features

  • Cross-Platform Support: Built on the Electron framework, compatible with Windows, macOS, and Linux.
  • Simple Configuration: Just enter your ASR API and AI API keys to use. Supports OpenAI-compatible providers (default is Alibaba Cloud DashScope).
  • Developer Friendly: Provides a complete developer mode, easy to extend and develop secondarily.
  • Privacy Secure: Configuration and history data are stored locally; APIs communicate directly with service providers.

🎬 Feature Demo

✨ Polish Mode

Enable AI Polishing: Long press Right Alt to speak, release to optimize expression and make the text smoother.

Polish Mode

⌨️ Command Mode

Select Text: Long press Right Alt and speak a command, release to precisely adjust text content to meet personalized needs.

Command Mode

💡 Ask Mode

Long press Right Ctrl and ask a question, release to get an AI answer.

Ask Mode

📸 Interface Showcase

Home Interface

Statistics

History

Settings - Model Configuration

Settings - Custom Prompt

📥 Download & Install

Voke provides installers for Windows, macOS, and Linux platforms.

  • Windows: Download
  • macOS: (Coming Soon)
  • Linux: (Coming Soon)

🛠️ Developer Mode

If you are a developer or want to experience the latest features, you can run from source:

1. Prerequisites

  • Node.js 18+
  • pnpm (Recommended) or npm

2. Get Source Code

git clone https://github.com/zyk/Voke.git
cd Voke

3. Install Dependencies

pnpm install

4. Start Development Mode

pnpm run dev

This command will start both the Electron main process and the Vite renderer process with hot reload support.

5. Build

# If encountering proxy bypass issues with app-builder
# For Windows, run the following commands first (7890 is your VPN port)
# $env:HTTP_PROXY="http://127.0.0.1:7890"
# $env:HTTPS_PROXY="http://127.0.0.1:7890"

# For macOS/Linux, run the following commands first
# export HTTP_PROXY="http://127.0.0.1:7890"
# export HTTPS_PROXY="http://127.0.0.1:7890"

# Build for specific platform
pnpm run build:win   # Windows
pnpm run build:mac   # macOS
pnpm run build:linux # Linux

⚙️ Quick Configuration

After starting the application, go to the Settings Page for quick configuration:

  1. ASR Configuration (Speech Recognition):
    • Enter your ASR API Key.
    • Default supports Alibaba Cloud DashScope (Qwen-ASR), also configurable for other OpenAI-compatible ASR services.
  2. AI Configuration (Large Model):
    • Enter your AI API Key.
    • Configure Base URL and Model Name (e.g., qwen-flash).

Tip: Configuration information is stored encrypted on your local device and will not be uploaded to any third-party servers.