# joewongjc/type4me

MacOS语音输入法，实时识别、大模型文本优化、全本地存储

Repository: https://github.com/joewongjc/type4me
Canonical: https://ross.abutalabs.com/products/type4me
Language: Swift
License: MIT
License Family: permissive
Last push: 2026-08-26T17:18:16+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 99, longevity 11
- inputs: {"age_days": 164, "days_push": 7, "days_rel": 7, "gap_med": 0, "n_releases_24m": 40}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1404, forks 165 (observed 2026-08-28T04:04:37.972240+00:00)

## What it is
Type4Me is a macOS voice input method that transcribes speech in real time (streaming, local or cloud ASR engines) and refines text with LLMs, with all history stored locally. It offers multiple modes like translation, smart polishing, Mac voice commands, and customizable hotword/mapping vocabularies.

## Use cases
- dictate text instead of typing on mac
- voice input with llm text cleanup
- translate speech to another language in real time
- control mac with voice commands
- fix just-dictated text by voice
- private local speech recognition
- custom hotwords for asr accuracy

## When to choose
- you want a fast, customizable, privacy-friendly voice input on macOS with local history export
- you need streaming dictation plus LLM polishing, translation, or voice-driven Mac actions
- you want to plug in your own ASR/LLM providers or run local models via Ollama

## When to avoid
- you need voice input on Windows or Linux
- you want a fully free offline solution without API keys and lack 32GB RAM for the local engine
- you need an input method for non-macOS or mobile platforms

## Facets
- artifact type: application
- maturity: active
- function: speech-recognition, nlp, llm-inference, tts
- domain: speech-processing, developer-tools, apple-ecosystem
- platform: -
- tags: voice-input, dictation, input-method, local-first, hotwords, translation, menu-bar-app, streaming-asr, natural-language-processing, macos, swift, desktop

## Member repositories
- joewongjc/type4me (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:37.972240+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 2026-08-30T04:38:50.980177+00:00, confidence not recorded.
  - readme: https://github.com/joewongjc/type4me (fetched 2026-08-28T04:04:37.972240+00:00, sha 93bfa364f221)
- Data as of 2026-08-30T08:39:29.467469+00:00.