v0.3.2

Please refer to the change log for a complete history. The main update includes minor bug fixes related to the upgrade process. Additionally, we have addressed a security issue concerning the storage of certain red-allowed text in the debugging log file.

v0.3.1

Fixed

• Hotkey now preserves the user's clipboard — original clipboard content is saved before Ctrl+C text capture and restored afterward

v0.3.0-beta

Fixed

• Speech permanently stops working after interrupting playback with a rapid double hotkey press (stop event was never cleared between speech attempts)
• Application crash on clean systems due to Python 3.14 NumPy native extension incompatibility
• Application crash on target systems caused by NumPy 2.x ABI incompatibility with onnxruntime (pinned NumPy < 2.0)
• Installer now installs to C:\Program Files\ instead of C:\Program Files (x86)\ on 64-bit systems

ReadToMe v0.2.0

The first major release has been in beta testing for nearly a week, and we have not encountered any significant issues. Everything appears to be functioning as expected.

The primary reason for the large download size is the inclusion of voice models. This version is pre-packaged with four different voice model options. For this reason, I have also created a portable light version that includes only one voice model, which is my preferred option.