VocalFold 1.3.0

Changelog

All notable changes to this project will be documented in this file.

1.3.0 - 2025-11-20

Added

• Executable Commands feature with application launcher functionality
• Usage tracking for commands and keyword replacements

[1.2.3] - 2025-11-06

Added

• Automatic Releases on GitHub when a new Version is tagged

1.1.0 - 2025-11-03

Added

• Microphone Settings page with device selection and testing capabilities
• Real-time audio level monitoring and visualization
• Microphone test functionality with transcription testing
• Changelog page to track version history
• External configuration file support for keywords (for GDrive/OneDrive/etc. sync)

Changed

• Improved settings UI with file selection for keywords path
• Enhanced AudioRecorder with level and spectrum callbacks
• Updated SystemSettings to integrate microphone configuration
• Reduced verbosity of Fast Fourier Transform log output

Fixed

• HTTP 500 error when saving settings changes
• Settings API error handling and response serialization

---

1.0.0 - 2025-11-01

Added

• Initial public release
• Voice-to-text transcription with Whisper
• Keyword replacement system with categories
• Hotkey support for quick voice input
• System tray integration
• Auto-start with Windows option
• General and system settings management

---

Microphone settings

[1.1.0] - 2025-11-03

Added

• Microphone Settings page with device selection and testing capabilities
• Real-time audio level monitoring and visualization
• Microphone test functionality with transcription testing
• Changelog page to track version history
• External configuration file support for keywords (for GDrive/OneDrive/etc. sync)

Changed

• Improved settings UI with file selection for keywords path
• Enhanced AudioRecorder with level and spectrum callbacks
• Updated SystemSettings to integrate microphone configuration
• Reduced verbosity of Fast Fourier Transform log output

Fixed

• HTTP 500 error when saving settings changes
• Settings API error handling and response serialization

---

VocalFold 1.0

VocalFold brings system-wide AI voice-to-text dictation to Windows — fully offline, privacy-first, and powered by Whisper.NET.

✨ Highlights

Global Hotkey Dictation – Press Ctrl+Windows, speak, and text appears instantly in any app.

Local AI Processing – Runs Whisper.NET entirely offline — no cloud, no tracking.

GPU Acceleration – Supports NVIDIA (CUDA), AMD (Vulkan), and Intel (Vulkan).

Modern Web UI – Configure settings, hotkeys, and keyword replacements via a clean web interface.

Keyword Shortcuts – Say “email signature” or “comma” and let VocalFold type it for you.

Always Ready – Runs in the system tray and can auto-start with Windows.

⚙️ Requirements

Windows 10 / 11

Optional GPU: NVIDIA RTX 20+, AMD RX 6000+, or Intel Arc

.NET 9.0 SDK (for building from source)

8–16GB RAM recommended

🚀 Installation

Option 1: Download the prebuilt executable from the Releases page.
Option 2: Build from source using .NET:

git clone https://github.com/yourusername/VocalFold.git
dotnet build
dotnet run --project VocalFold
🧠 Benchmarks (5s speech, Base model)

RTX 3080: <1s

RX 6800 XT: <1.5s

Intel Arc A750: <2.5s

CPU: 5–8s

🔒 Privacy First

All audio is processed locally — no cloud, no telemetry, no storage. Once the model downloads, VocalFold works entirely offline.

💙 Support Development

Enjoying VocalFold?
☕ Buy me a coffee on Ko-fi — your support keeps this free, open-source project alive!

Built with F# + Whisper.NET + React + TailwindCSS
Made with ❤️ by the VocalFold team