Version 1.12

Updated models source URL in documentation and GUI
Reliability enhancement, microphone capture less likely to transition to “Stalled” state and discard the audio

Version 1.11

When loading models, adapters can be selected with 0-based index, in addition to the name.
Added an API method to decode initial prompt into array of tokens.

Version 1.10.1

Transcribe-File PowerShell command now supports human-readable language names
Better error messages when the library is unable to decode input audio files

Version 1.10

New API method to decode audio files from memory buffers.
The library should now support concurrent transcribes, when loading multiple models.
Added an API method to clone the model, and the corresponding flag in the eGpuModelFlags enumeration. The cloned model uses same GPU and parameters, reuses VRAM buffers with model’s tensors, but it has independent D3D device and context.

The desktop app now has “Stop” button while transcribing files.

Created PowerShell 5.1 wrapper, see the readme in the corresponding folder.

Version 1.9

Removed the dependency on cabinet.dll Windows component.
Reportedly, the library now runs on Linux, using Wine 8.3.

Version 1.8.2

Added new command-line arguments to the C++ console app, to select which GPU to use for the inference.

Version 1.8.1

Fixed a crash during process shutdown

Version 1.8

• Changed library API to allow to select GPU for the inference
• Added another combobox on the advanced GPU settings dialog, to manually select the GPU
• Documented F16C requirement, and improved error handling accordingly

Also upgraded the compilers, this release is built with Visual Studio 2022 version 17.5.1

Version 1.7

Bugfixes, API documentation

Version 1.6.1

Minor changes in the desktop app, the DLL is still 1.6.0

• Better performance of C++ samples on laptops with two graphics cards
• Added *.m4a file extension to the browse dialog
• “Text with timestamps” output format option