EchoType / 声笔

🎤 Fast · Free · Offline Voice Input Tool
🎤 快速 · 免费 · 离线的语音输入工具

Languages / 语言: English | 中文

📥 Download Latest · 📖 Documentation · 🐛 Issues
📥 下载最新版本 · 📖 使用文档 · 🐛 问题反馈

---

📖 Project Overview

EchoType / 声笔 is an intelligent voice input tool designed specifically for Windows, enabling you to quickly convert speech to text input. Built upon and optimized from the CapsWriter-Offline project, it provides a completely offline voice recognition experience.

🎯 Core Advantages

• 🔒 Privacy & Security: Completely offline processing, no data upload
• ⚡ Lightning Fast: Local models with millisecond-level recognition
• 🎨 Simple & Easy: System tray resident, one-click activation
• 🛠️ Highly Customizable: Rich personalization settings

✨ Features

🎤 Voice Recognition

• Real-time Conversion: Instant speech-to-text with rapid response
• High Accuracy: Based on advanced local AI models
• Multiple Input Modes: Support for press-and-hold and click-to-toggle recording
• Custom Hotwords: Add professional vocabulary to improve recognition accuracy

🖥️ User Interface

• System Tray Resident: Real-time status display (idle/connected/recording/error)
• Friendly Tooltips: Intuitive hover information
• Graphical Settings: Intuitive configuration interface
• Hotkey Support: Customizable shortcuts for quick activation

🔧 System Integration

• Auto-start: Optional Windows startup item
• Audio Devices: Smart microphone detection and switching
• Output Formats: Multiple text format options
• Bubble Notifications: Real-time desktop feedback

📊 Management Tools

• Logging System: Detailed runtime log recording
• Error Diagnostics: Convenient troubleshooting tools
• Configuration Backup: Settings import/export functionality

🚀 Quick Start

📥 Installation Steps

1. Download Program

   Download the latest EchoType.exe from releases
   

   👉 Click to Download

2. Launch Program

   Double-click EchoType.exe to run
   Program icon will appear in system tray
   

3. Configure Settings

   Right-click tray icon → Settings
   Configure hotkeys and personal preferences
   

4. Start Using

   Press the configured hotkey
   Begin voice input experience
   

⌨️ Default Hotkeys

• F4: Start/Stop recording
• Right-click tray: Open menu
• Double-click tray: Quick settings

📋 System Requirements

Item	Requirement
Operating System	Windows 10/11 (64-bit)
Memory	Minimum 4GB RAM
Storage	At least 2GB available space
Audio Device	Supported microphone device
Network	No internet connection required

🔧 FAQ

❓ Program won't start?

1. Check if antivirus software is blocking it
2. Confirm Windows version compatibility
3. Check log files for error diagnosis
4. Try running as administrator

❓ Voice recognition inaccurate?

1. Check microphone device and volume
2. Use in quiet environment
3. Add custom hotwords to improve accuracy
4. Adjust recording sensitivity settings

❓ How to add custom hotwords?

1. Right-click tray icon and select "Settings"
2. Go to "Hotword Management" tab
3. Add frequently used professional vocabulary
4. Save settings and restart program

❓ Installation issues with sherpa-onnx?

Problem: ModuleNotFoundError: No module named 'cmake.cmake_extension' or compilation errors

Solution:

1. Use precompiled packages instead of building from source:

   pip install --find-links https://k2-fsa.github.io/sherpa/onnx/install/python.html sherpa-onnx
   pip install funasr-onnx==0.2.5

2. Make sure Visual Studio Build Tools with C++ support is installed
3. Install cmake: pip install cmake
4. If still failing, use client-only mode and connect to a remote server

🛠️ Development & Packaging

🔧 Developer Guide

Architecture

EchoType uses a client-server architecture:

• Client (run_tray.py): Tray icon, hotkey monitoring, audio recording
• Server (server/): Voice recognition service (requires sherpa-onnx, etc.)

Environment Setup (Full, with Local Server)

The voice recognition server requires sherpa-onnx, funasr-onnx, etc.

Prerequisites:

1. Install Visual Studio Build Tools with "Desktop development with C++".
2. Install CMake.

Installation Steps:

# 1. Create and activate a virtual environment
python -m venv .venv
.venv\Scripts\activate

# 2. Install dependencies
pip install -r requirements.txt

Running in Development

# 1. Start the server in the background
start /B python server/start_server.py

# 2. Run the client
pythonw run_tray.py

📦 Packaging for Distribution

To package the project into standalone executables, PyInstaller is recommended.

Prerequisites

1. Setup Virtual Environment:

   python -m venv .venv
   .venv\Scripts\activate
   pip install -r requirements-simple.txt
   pip install --find-links https://k2-fsa.github.io/sherpa/onnx/install/python.html sherpa-onnx
   pip install funasr-onnx==0.2.5
   pip install kaldi-native-fbank

2. Install PyInstaller:

   pip install pyinstaller

Build Process

Step 1: Build All Components

# Build client
call .venv\Scripts\pyinstaller.exe EchoType.spec

# Build server
call .venv\Scripts\pyinstaller.exe EchoTypeServer.spec

# Build server manager
call .venv\Scripts\pyinstaller.exe EchoTypeServerManager.spec

Step 2: Merge Distribution

call .venv\Scripts\python.exe build_package_en.py

Step 3: Test Package

call .venv\Scripts\python.exe test_package.py

Final Package Structure

EchoType_Release/
├── EchoType.exe              (Main client app)
├── EchoTypeServer.exe        (Backend server)
├── EchoTypeServerManager.exe (Server Manager UI)
└── _internal/
    ├── assets/               (UI resources)
    ├── hotwords/             (Custom vocabulary)
    ├── locales/              (Translations)
    ├── models/               (AI models)
    └── (all dependencies)

Notes

• The .spec files are pre-configured with correct paths and dependencies
• build_package.py automatically merges all three dist folders
• See PACKAGING_GUIDE.md for troubleshooting

---

🌟 If this project helps you, please give it a Star!

⭐ Star · 🍴 Fork · 📝 Issues

---

Made with ❤️ by ljyou001