# Const-me/Whisper

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

Repository: https://github.com/Const-me/Whisper
Canonical: https://ross.abutalabs.com/products/const-me-whisper
Language: C++
License: MPL-2.0
License Family: copyleft
Last push: 2026-05-24T09:42:55+00:00

## Health v2 (maintenance only)
Score: 60/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 84, release rhythm 8, longevity 95
- inputs: {"age_days": 1334, "days_push": 101, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 10649, forks 965 (observed 2026-08-28T04:10:43.647525+00:00)

## What it is
A Windows port of whisper.cpp that runs OpenAI's Whisper speech recognition model on the GPU via DirectCompute (Direct3D 11 compute shaders). It ships as a small standalone Whisper.dll with a COM-style API plus a WhisperDesktop GUI for transcribing files and live microphone audio.

## Use cases
- transcribe audio files to text on windows
- real-time speech to text from microphone
- run whisper faster than pytorch on gpu
- offline speech recognition without python dependencies
- integrate whisper asr into a c# application
- translate foreign-language audio to english

## When to choose
- you need fast GPU-accelerated Whisper inference on Windows with minimal dependencies
- you want a desktop app to transcribe files or live audio without installing Python or CUDA
- you need a compact native library (COM API / C# wrapper) to embed ASR in your own software

## When to avoid
- you need Linux or macOS support - this is Windows-only
- you need Ogg Vorbis input or ASIO-only capture devices
- you want the actively developed cross-platform whisper.cpp or faster-whisper ecosystem instead

## Facets
- artifact type: application
- maturity: maintenance
- function: speech-recognition, llm-inference, gpu-computing, audio-processing
- domain: speech-processing, machine-learning, windows
- platform: windows, cpp
- tags: whisper, asr, directcompute, transcription, gpgpu, voice-activity-detection, whisperdesktop, gpu, desktop

## Member repositories
- Const-me/Whisper (main) score 60

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:43.647525+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:18:13.402143+00:00, confidence not recorded.
  - readme: https://github.com/Const-me/Whisper (fetched 2026-08-28T04:10:43.647525+00:00, sha 900ae64787ed)
- Data as of 2026-08-30T08:39:29.467469+00:00.