xiaohuailabs/xiaohu-video-translate — explained in plain English
Analysis updated 2026-05-18
Turn a YouTube video in any language into a video with burned-in Chinese subtitles using one plain-language request.
Generate a clean Chinese transcript from a foreign-language talk without adding subtitles to the video.
Add Chinese subtitles to a local video file already saved on your computer.
Produce bilingual Chinese and English subtitles for language learning.
| xiaohuailabs/xiaohu-video-translate | joeseesun/qiaomu-goal-meta-skill | patchfighterway90/cs2-external-overlay | |
|---|---|---|---|
| Stars | 495 | 494 | 494 |
| Language | Python | Python | Python |
| Setup difficulty | — | easy | easy |
| Complexity | — | 2/5 | 2/5 |
| Audience | vibe coder | vibe coder | developer |
Figures from each repo's GitHub metadata at analysis time.
This project automatically adds Chinese subtitles to foreign-language videos. With a single plain-language request, it downloads a video, transcribes the speech, translates it into Chinese, cleans up the result, and burns the subtitles into the video. It also produces a clean written transcript. Everything runs on your own computer, and the transcription step costs nothing in service fees. The whole process is a five-step pipeline that runs by itself: download the video, extract the audio and transcribe it, translate it, polish the wording, and finally burn the subtitles onto the video. You just describe what you want, and it walks through each step. Any language that the transcription engine can understand, such as English, Japanese, Korean, French, or Spanish, can be turned into Chinese subtitles. Chinese videos are only transcribed, not translated. Subtitles can be Chinese only or Chinese and English together. Technically, the project is a set of three skills, meaning script-and-instruction bundles that AI coding command-line tools can run. The skills are not tied to one tool: Claude Code, OpenClaw, Gemini CLI, and Codex can all drive them. One skill acts as the overall coordinator, one handles subtitle translation and polishing, and one is a pure downloader. The three can be used together or separately. Transcription uses Whisper, an open speech-recognition system, with extra speed on Apple Silicon Macs. The translation step reuses whatever AI you already have, so you do not need to pay for a separate translation service. The author stresses accuracy: word-level timing so subtitles do not run ahead of the speaker, automatic correction of misheard technical terms, and sentence breaks based on meaning. Installation differs by operating system. On macOS you install a couple of base tools, a Whisper engine, then clone the repository and run an install script. On Windows and Linux there are specific differences, especially the need to switch the subtitle font to one that supports Chinese, since the default is a Mac-only font. After setup you restart your AI coding tool and speak naturally to it. The README also covers common problems, such as download errors and missing fonts, and warns against committing your personal configuration. It is released under the MIT License.
AI coding agent skills that auto-download, transcribe, translate, and burn Chinese subtitles onto foreign-language videos.
Mainly Python. The stack also includes Python, Whisper, yt-dlp.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.