hey! this is a static git repo viewer. commit diffs, file history, older commits and other revisions won't load without javascript.

you are still able to view source files on branch tips, but the experience is degraded.

yum-archive/TaSTT-Whisper

High-performance GPGPU inference of OpenAI's Whisper automatic speech recognition (ASR) model

git clone https://git.yummers.dev/yum-archive/TaSTT-Whisper

yumFix audio normalization59297502023-04-04

files

API/
Internal/
Callbacks.cs	1.0 KiB
CaptureCallbacks.cs	1.1 KiB
Context.cs	7.4 KiB
ExtensionMethods.cs	2.1 KiB
Library.cs	4.5 KiB
Readme.md	883 B
Readme.txt	78 B
WhisperNet.csproj	955 B
WhisperNet.nuspec	1.2 KiB

This library implements high-performance GPGPU inference of OpenAI's Whisper automatic speech recognition (ASR) model.

The library requires a hardware GPU which supports Direct3D 11.0, a 64-bit Windows OS, only works within 64-bit processes, and requires a 64 bit CPU which supports SSE 4.1.

The main entry point of the llibrary is Whisper.Library static class. Call loadModel function from that class to load an ML model from a binary file.

These binary files are available for free download on Hugging Face. I recommend ggml-medium.bin (1.42GB in size), because I’ve mostly tested the software with that model.

Once the model is loaded, create a context by calling createContext extension method, then use that object to transcribe or translate multimedia files or realtime audio captured from microphones.