Version 2024.05.22

New in version 2024.05.22:

• Update Turkish translation
• Fix untranslatable strings
• Remove redundant image processing code, only read the cache if that option is selected

version 2024.05.16

New in this version:

• Fixed a critical bug with the dependencies where in extremely limited circumstances, the addon would cause NVDA to continuously restart upon crashing due to a missing DLL file with the face detection interface.
• We now support the new GPT4O (GPT4 omni) model announced at OpenAI dev day. This model boasts significant speed improvements over its predecessors.
• Updated Russian localization
• Ukrainian translation updates
• Add turkish translation
• There are now more strings, everything in the addon should be able to be translated now including the facial position results. A special thank you to all the translators!

Version 2024.04.14

New in this release:

• Adds support for NVDA 2024.1 and python 3.11
• Removes dependencies from the addon package to reduce overall size and footprint. Now, following installation and NVDA restart, a dialog will pop up if the needed libraries are unavailable. This effectively means we can support any python upgrades that NVDA might make without having to distribute another package.

libs

The external third-party dependencies used by this addon. The .zip files should be replaced directly.

When something breaks with package imports, the code will try to pull these in automatically.

Version 2024.03.29

New in this version:

• Added face detection, which will tell you whether you are clearly centered into the frame of your camera. The hotkey NVDA+shift+j is bound by default, and options to select a different device or release the camera so that other apps can use it may be found in the AI content describer context menu.
• Added an option and script to take a picture from the onboard webcam.
• Added Google's Gemini model.
• Added the three major Claude 3 models (Haiku, Sonnet, and Opus).
• Added support for Llama.cpp.
• Rewrote the AI content describer section of the NVDA settings dialog, making many of preferences model specific. On installation of the new version your settings should automatically be ported over.
• Added a model selection submenu to the AI content description context menu.
• Changed some default options to make them more logical.
• Now, when you trigger a description action, the model in use will be spoken.

Version 2024.03.13

New in this version:

• Fixes an urgent bug that could have prevented the addon from working with certain braille display models.

Version 2024.03.09

New in this version:

• Support descriptions formatted in HTML and Markdown, thanks Mazen)
• Case is no longer taken into account when deciding whether a file extension is valid
• for developers: When the add-on is running from scratchpad and translations are unavailable, silently log instead of showing an error
• Documentation for translating the project
• Russian translation, thanks Kostenkov
• Ukrainian translation, thanks Heorhii Halas
• Serbian translation, thanks nidza07

Version 2023.11.23

New in this version:

• Adds PYD files to support python 3.11 and NVDA 2024.1 (not guaranteed to be stable as the API isn't frozen yet)
• Adds a copyright header with license information to each source file
• Released early because we need a build number that is supported by the add-on store

Version 2023.11.22.1

New in this version:

• Object detection should now work just about everywhere, including in the browser through the menu and other contexts
• Error when we try to snap something while screen curtain is enabled

Version 2023.11.22

this backward incompatible version fixes a few bugs:

• The menu now shows up as one might expect. Strange concept!
• Internal name changed from "AI Content Describer" to "AIContentDescriber" to honor manifest conventions. Normally this would mean multiple versions running side-by-side, but there is now a task that will delete the older one if it detects it during installation.