# matthartman/ghost-pepper

100% private on-device voice models for speech-to-text and meeting transcription on macOS

Repository: https://github.com/matthartman/ghost-pepper
Canonical: https://ross.abutalabs.com/products/ghost-pepper
Language: Swift
License Family: other
Last push: 2026-07-27T14:25:47+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 94, release rhythm 95, longevity 11
- inputs: {"age_days": 166, "days_push": 37, "days_rel": 37, "gap_med": 1.0, "n_releases_24m": 21}
- flags: young, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3140, forks 187 (observed 2026-08-28T04:07:45.846255+00:00)

## What it is
A free, open-source macOS menu bar app providing fully on-device speech-to-text dictation and meeting transcription using local Whisper, Parakeet, and Qwen ASR models. It includes a local LLM cleanup step that removes filler words and saves meeting transcripts and AI summaries as markdown, with no data leaving the machine.

## Use cases
- transcribe voice to text locally on mac without cloud
- private meeting transcription with summaries
- hold a hotkey to dictate into any app
- offline dictation for apple silicon mac
- convert meeting recordings to markdown notes
- speech to text in 50+ languages on device

## When to choose
- you need private, on-device dictation with no cloud APIs
- you want meeting transcripts and AI summaries stored locally as markdown
- you're on an Apple Silicon Mac running macOS 14+
- you want multilingual speech recognition without sending audio anywhere

## When to avoid
- you need Windows or Linux support
- you're on an Intel Mac or macOS 13 or older
- you need cloud-scale batch transcription of large audio archives
- you need speaker diarization or team collaboration features

## Facets
- artifact type: application
- maturity: active
- function: speech-recognition, nlp, llm-inference, audio-processing
- domain: speech-processing, privacy, developer-tools, apple-ecosystem
- platform: -
- tags: dictation, meeting-transcription, on-device-ai, whisper, menu-bar-app, apple-silicon, local-llm, macos, desktop, swift

## Member repositories
- matthartman/ghost-pepper (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:45.846255+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 2026-08-30T07:26:02.355043+00:00, confidence not recorded.
  - readme: https://github.com/matthartman/ghost-pepper (fetched 2026-08-28T04:07:45.846255+00:00, sha 53b18a85a923)
- Data as of 2026-08-30T08:39:29.467469+00:00.