Fox Reader is a simple text-to-speech application built in Rust and GTK4 that converts text to speech using high-quality voices from Kokoros TTS.

1. AI Chat with LLM Integration

   • Use local LLM solutions (Ollama/LM Studio)
   • Connect to AI models via API keys (OpenAI, etc.)
   • Voice-to-text capability using Whisper models

2. PDF Reader with Highlighting System

   • Simple PDF render system
   • Read PDF documents with real-time text highlighting
   • Choose from where to start reading

3. Text-to-Speech with Highlighting System

   • Convert any text to natural-sounding speech using Kokoros voices

4. Speech Dispatcher Compatibility

   • Seamless integration with Linux accessibility tools
   • Works with system-wide speech services

5. Firefox Read Aloud Integration

   • Works directly with Firefox's built-in reading feature
   • Use voices downloaded via app

• Operating Systems: Primarily Linux-based distributions, tested on:

  • Ubuntu 24.04
  • Fedora 42
  • Arch Linux (GNOME & Hyprland)

• GTK4 ≥ 4.12, Check version with:

pkg-config --modversion gtk4

• Adwaita >= 1.5, check version via

pkg-config --modversion libadwaita-1

• Speech Dispatcher for reading via spd-say or browser (optional), example install on Arch Linux:

sudo pacman -S speech-dispatcher

• Rust toolchain (for building from source) Install rust

• On Ubuntu 24.04, Fedora 42, and Arch Linux with GNOME, no extra dependencies are needed—required GTK4 and Adwaita libraries are already included.

• On Hyprland (Arch Linux), libadwaita is not installed by default. You must install it manually:

sudo pacman -S libadwaita

• Older versions of Ubuntu, Debian, Fedora etc. can have problem with running this as updating gtk4 to newer version can be a bit difficult.

1. Download the latest release:

curl -sL $(curl -s https://api.github.com/repos/kopecmaciej/fox-reader/releases/latest \
  | grep browser_download_url \
  | grep .tar.gz \
  | cut -d '"' -f 4) -o fox-reader.tar.gz

tar -xzf fox-reader.tar.gz
## Or other desired path
mv fox-reader ~/.local/bin/
rm fox-reader.tar.gz

2. Run the application:

fox-reader

1. Clone the repository:

# Clone repository
git clone https://github.com/kopecmaciej/fox-reader.git
cd fox-reader

# Build
cargo build --release

# Run
./target/release/fox-reader

1. Open the Voice List tab
2. Browse available voices from the Kokoros TTS repository
3. Download desired voices; every downloaded voice is available in PDF Reader, Text Reader, and Speech Dispatcher
4. Set favorite voice as default for Speech Dispatcher usage without specified model

Fox Reader integrates with Speech Dispatcher through the app's CLI mode. A special script located in ~/.config/speech-dispatcher/fox-reader.sh will forward data and options properly. If Fox Reader is missing in $PATH you have to specify location by yourself in the script.

Fox Reader can be used via command line interface for quick text-to-speech conversion without launching the GUI.

fox-reader --cli --model <MODEL_PATH> --text <TEXT> [--rate <RATE>] [--output <OUTPUT_PATH>]

• --cli: Run in CLI mode without launching the GUI
• --text or -t: Text to synthesize

• --voice or -v: Voice name from Voice List tab, example: pm_alex
• --speed or -s: Speech rate adjustment (0.5 to 2)
• --output or -o: Path to save the audio output in WAV format
  • If not specified, audio will play immediately
• --list-voices or -l: List all available voices

Play speech immediately:

fox-reader --cli --voice pm_alex --text "Hello, this is Fox Reader speaking."

Adjust speech rate:

fox-reader --cli --voice pm_alex --text "This is faster speech." --speed 1.5

Save to file instead of playing:

fox-reader --cli --voice pm_alex --text "This will be saved to a file." --output ~/output.wav

Fox Reader uses GSettings for storing user preferences and configuration options. These settings include:

• UI theme and appearance preferences
• Default speech parameters
• Window size and position
• Selected voice preferences
• AI chat configuration (model selection, api keys, temperature settings)

You can view and modify these settings using the built-in preferences dialog or through the gsettings command-line tool. All other assets like voices, the pdfium library, and whisper models are stored separately in ~/.local/share/fox-reader/.

1. Voice download fails

   • Check your internet connection
   • Ensure you have write permissions to the voices directory

2. PDF reader doesn't load

   • Make sure pdfium is installed or let Fox Reader install it automatically
   • Check if the PDF file is not corrupted or password-protected

3. Speech Dispatcher integration issues

   • Verify Speech Dispatcher is installed and running
   • Check the configuration in ~/.config/speech-dispatcher/

• Improve performance for better user experience on slow PCs
• Enhance the PDF reader with better text extraction
• Optimize voice processing for lower latency

• Experiment with higher-quality voice models
• Consider migration to another GUI Rust library for better cross-platform support
• Expand AI chat functionality

Contributions are welcome! Please feel free to submit a Pull Request.

• Kokoros TTS for providing high-quality TTS voices
• GTK4 team for the UI framework
• Speech Dispatcher project
• Pdfium-render After using multiple crates that work with PDFs, this one seems the best
• Whisper-rs seamless integration with Whisper models.