whatisgithub

What is whicc?

nbzz/whicc — explained in plain English

Analysis updated 2026-05-18

35SwiftAudience · generalComplexity · 3/5LicenseSetup · moderate

In one sentence

A macOS app that overlays real time translated subtitles on any video or livestream using local speech recognition and a translation model you set up yourself.

Mindmap

mindmap
  root((whicc))
    What it does
      Live subtitle overlay
      Local speech recognition
      Real time translation
    Tech stack
      Swift
      SwiftUI
      Python
      Apple Silicon MLX
    Use cases
      Foreign livestreams
      Lectures abroad
      AI interviews
    Audience
      Mac users
      Language learners

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Watch a foreign language livestream with live translated subtitles

USE CASE 2

Follow a lecture in a foreign language using microphone mode

USE CASE 3

Keep captions on screen while watching an AI interview in another language

USE CASE 4

Study a language by comparing bilingual subtitles in real time

What is it built with?

SwiftPythonSwiftUIMLX

How does it compare?

nbzz/whiccrepoprompt/repoprompt-classiczaidmukaddam/swift-ai-sdk
Stars353535
LanguageSwiftSwiftSwift
Setup difficultymoderatemoderatemoderate
Complexity3/53/53/5
Audiencegeneraldeveloperdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Requires installing and configuring a separate LM Studio translation server, without it, subtitles are not translated.

MIT license lets you use, copy, modify, and sell the app freely as long as you keep the original copyright notice.

So what is it?

Whicc is a macOS app that shows real time translated subtitles over any video you are watching, whether that is a livestream, a recorded video, or even sound picked up from your microphone during a lecture. The idea is not to be a perfect translator, but to always be there so you can follow along in a language you do not fully understand. The app works by running speech recognition locally on your Mac using Apple Silicon, so the audio itself never leaves your machine. It uses one recognition model tuned for Chinese and another tuned for other languages. Once the speech is turned into text, that text is sent to a separate translation model that you set up yourself using a free program called LM Studio, which can run on the same Mac or on another computer on your home network. Whicc does not include or ship a translation model itself. Without a translation service configured, the app still shows live recognized text, but it will not produce a finished translated caption. Subtitles appear in a small floating window that stays on top of other apps and can be repositioned. The app also builds a private glossary of terms over time, learning from the content you watch so translations of specific vocabulary improve in later sessions. Requirements are macOS 15 or newer running on Apple Silicon, an internet connection for the first download of the speech models, and a separate machine or the same Mac to run the translation service. Normal users can download a ready made app from the releases page, while developers can build it from source following the project's development guide. The project is released under the MIT license.

Copy-paste prompts

Prompt 1
Walk me through installing whicc and setting up LM Studio as its translation service.
Prompt 2
Explain how whicc keeps my audio private while still translating in real time.
Prompt 3
Help me configure whicc to use a translation model running on another computer on my network.
Prompt 4
Show me how whicc's self learning glossary improves translations over time.

Frequently asked questions

What is whicc?

A macOS app that overlays real time translated subtitles on any video or livestream using local speech recognition and a translation model you set up yourself.

What language is whicc written in?

Mainly Swift. The stack also includes Swift, Python, SwiftUI.

What license does whicc use?

MIT license lets you use, copy, modify, and sell the app freely as long as you keep the original copyright notice.

How hard is whicc to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is whicc for?

Mainly general.

Open on GitHub → Ask about another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.