EchoType

EchoType — a client-server application for Speech-to-Text based on faster-whisper and inspired by AquaVoice.

Features

• 🎤 Voice recording via hotkeys — Push-to-Talk and Toggle modes
• 🚀 Fast recognition — based on faster-whisper with GPU (CUDA) support
• 📋 Text insertion — automatic typing into active application or clipboard copy
• 🖼️ GUI interface — system tray applet, popup window with audio visualization, settings window
• ⚙️ Flexible configuration — YAML-based configuration

Architecture

flowchart TB
    subgraph GUI[GUIClient]
        TRAY[Tray Applet]
        POPUP[Popup Window]
        SETTINGS[Settings Window]
        
        subgraph Core[Client - Core Logic]
            HK[HotkeyManager]
            REC[AudioRecorder]
            HTTP[HTTP Client]
            ACTION[Action]
        end
    end
    
    subgraph Server[STT Server]
        WH[WhisperModel]
        API[REST API]
    end
    
    subgraph Config[Configuration]
        CM[ConfigManager]
        YAML[config.yaml]
    end
    
    TRAY --> POPUP
    TRAY --> SETTINGS
    TRAY --> Core
    POPUP --> Core
    
    HK -->|triggers| REC
    REC -->|audio data| HTTP
    HTTP -->|POST /transcribe| API
    API --> WH

    WH -->|transcription|ACTION
    
    CM --> YAML
    CM --> GUI
    CM --> Server

Loading

Components

Component	File	Description
STTServer	STTServer/stt_server.py	FastAPI server with Whisper model
GUIClient	GUIClient/gui_client.py	PyQt6-based GUI client, uses Client internally
Client	Client/client.py	Core client logic, coordinates HotkeyManager, AudioRecorder and server communication
HotkeyManager	Client/HotkeyManager/	Hotkey management with PTT and Toggle modes
AudioRecorder	Client/AudioRecorder/	Audio recording from microphone
ConfigManager	config_manager.py	Singleton configuration manager

Project Structure

EchoType/
├── main.py                    # Server entry point
├── gui_client.py              # GUI client entry point
├── cli_client.py              # CLI client entry point
├── config.yaml                # Configuration file
├── config_manager.py          # Configuration manager
│
├── STTServer/                 # Speech-to-Text server
│   ├── __init__.py
│   └── stt_server.py
│
├── Client/                    # Client core
│   ├── __init__.py
│   ├── client.py
│   ├── AudioRecorder/         # Audio recording module
│   │   ├── audio_recorder.py
│   │   ├── audio_data.py
│   │   └── recording_state.py
│   └── HotkeyManager/         # Hotkey module
│       ├── hotkey_manager.py
│       ├── hotkey_action.py
│       ├── hotkey_mode.py
│       └── hotkey_state.py
│
├── GUIClient/                 # GUI components
│   ├── gui_client.py
│   ├── TrayApp/               # System tray applet
│   ├── Windows/               # Windows (popup, settings)
│   ├── Widgets/               # Widgets (visualizer, timer)
│   ├── Style/                 # QSS styles
│   └── SFX/                   # Sound effects
│
└── plans/                     # Documentation and plans

Tech Stack

Category	Technology
Server	FastAPI, uvicorn
STT	faster-whisper
GUI	PyQt6
Audio	sounddevice, soundfile
Hotkeys	pynput
Configuration	PyYAML

Quick Start

Installation

# Clone the repository
git clone https://github.com/Protectore/EchoType.git
cd EchoType

# Install dependencies (requires uv)
uv sync

Running

# Start the server
uv run main.py

# Start the GUI client (in another terminal)
uv run gui_client.py

Configuration

Main settings in config.yaml:

# Whisper model
model:
  size: medium        # tiny, base, small, medium, large-v3
  device: cuda        # cuda or cpu
  compute_type: float16

# Hotkeys
hotkeys:
  record:
    keys: alt_gr      # Recording key
    mode: ptt         # ptt (Push-to-Talk) or toggle

# GUI
gui:
  show_popup: true

Recording Modes

Push-to-Talk (PTT)

Hold the key to record. Release to stop recording and send for recognition.

Toggle

Press the key to start recording. Press again to stop.

Requirements

• Python 3.13+
• CUDA (optional, for GPU acceleration)

License

MIT

Support

You can support project/developer on DonationAlerts. Thank you!