philipag

Works with WINE on Linux/Mac thanks to using DirectX 11 instead of 12

12

For anyone interested: I have tested this library running an app under WINE 8.3 (www.winehq.org) on Linux and it works thanks to NOT requiring DirectX12 (WINE only goes up to...

running multiple concurrent threads (each with their own model+context) throws exceptions

4

I tried running 2 threads - each allocating its own model and context - but calling context.runFull() from both threads at the same time causes exceptions. Is this tested/supported? My...

Without seeking the interface behaves identically to before. This is very useful when wanting to read a stream in a non-linear fashion. If block 0 is not read first (required...

"Max attribute characters per line" is not respected even when "Attribute Tolerance"=5 and "Max attributes per line"=8

5

**Describe the bug** The MACPL setting says that it will work with AT=5 and MAPL=8, but it has no effect. If I set MACPL=20, then the attributes do not wrap...

Feature request: create notification which does not close when clicked

3

In Android I use notification which do not disappear when clicked, and I cannot replace those with this Nuget. Specifically this is the issue in the code: .SetAutoCancel(true) It would...

Document frame rate with your 128x96 panel.

4

Great detail in your readme but would be nice to know what frame rates you achieve with your panels.

Using the sample audio `whisper-cli -f samples/jfk.wav --grammar grammar.txt --grammar-rule "root" --grammar-penalty 50` with the attached grammar file does not give any output. Am I doing something obviously wrong here?...

Add C language binding for grammar-parser to core.

1

The main idea is to create a C facade for grammar loading so that it can easily be used to load/allocate (and later free) grammars via any type of language...

ggml_vulkan crashes if running 2 threads which allocate their own context for the same GPU device

3

I am using Whisper.net which in turn uses whisper.cpp which in turn uses ggml. Running a single thread with the Vulkan backend works perfectly (but does not saturate the GPU...

**Is your feature request related to a problem? Please describe.** ROCm Hip support cannot be built even though it is fully supported by ggml. The current build system makes it...