YouDub WebUI

// Overview

YouDub WebUI is an open-source video localization tool that takes a single YouTube, Bilibili or local video and produces a target-language version of it. You import the video, it recognises and translates the content, and then — depending on the task you pick — it outputs a hard-subtitled video that keeps the original audio, a dubbed video with no burned-in subtitles, or a video carrying both. In dubbing mode it separates the vocal track from the background, generates the dubbed speech and mixes the result, so music and sound effects survive the process. The finished video plays and downloads from the web UI.

Its most mature path is English-to-Chinese dubbing of YouTube material; Chinese-to-English dubbing of Bilibili videos is also supported, and Japanese-to-Chinese has been wired up for local files. The maintainer is candid about the state of each direction — the Japanese-to-Chinese path has passed its automated parameter and regression tests but has not yet been signed off against real Japanese media.

The project is not a demo. Its author runs a Bilibili channel with over a million followers and tens of thousands of videos, all of them automatically translated and dubbed with this tool across technology, gaming, science, animals and history. The stated goal is to let individual creators, developers and small teams own a complete localization pipeline locally while keeping the architecture simple enough to understand, debug and extend. It is a self-hosted stack rather than a one-click app: a Python 3.12 backend and a Node frontend, with FFmpeg, large ASR/TTS/separation models and — for full processing — a CUDA GPU. Authentication is mandatory, and the backend refuses to start without a password hash.

// What it does

• Full pipeline in one tool: download or import, transcribe, translate the subtitles, synthesise the dub, mix and render
• Three output modes per task — hard-subtitled with original audio, dubbed without burned-in subtitles, or both together
• Vocal and background separation via a bundled Demucs submodule, so music and sound effects are preserved under the dub
• Voice cloning so the dubbed track resembles the original speaker rather than a stock TTS voice
• Sources include YouTube, Bilibili and local video files, with proxy and cookie configuration for YouTube
• Any OpenAI-compatible Chat Completions endpoint drives the translation step — base URL, key and model are yours to choose
• Mandatory Argon2id login, with proxy settings, cookies, model cache and work folders all kept on the local machine

// Getting started

YouDub runs as a FastAPI backend plus a Next.js frontend. You need Python 3.12, Node.js 20+, FFmpeg and ffprobe on PATH, and an OpenAI-compatible endpoint for translation; a CUDA GPU is recommended because transcription, separation and TTS are very slow on CPU. The first run downloads large ASR, TTS and audio models.

# Ubuntu / Debian / WSL2
sudo apt update
sudo apt install -y ffmpeg nodejs npm

# macOS (Homebrew)
brew install ffmpeg node

git clone https://github.com/liuzhao1225/YouDub-webui.git
cd YouDub-webui
git submodule update --init --recursive

python3.12 -m venv .venv
.venv/bin/python -m pip install -U pip
.venv/bin/pip install -r requirements.txt

(cd apps/web && npm ci)

cp env.txt.example .env

.venv/bin/python -c "from getpass import getpass; from pwdlib import PasswordHash; print(PasswordHash.recommended().hash(getpass('YouDub password: ')))"

.venv/bin/uvicorn backend.app.main:app --reload --host 0.0.0.0 --port 8000

npm --prefix apps/web run dev -- --hostname 0.0.0.0 --port 3000

Commands and code are distilled from the project's own documentation — always check the official repo for the latest.

// When to use it

• Republish your own videos in another language with a cloned voice instead of commissioning a dub per video
• Batch-localize a back catalogue of technical or educational content for a second-language audience
• Produce translated hard subtitles while keeping the original audio, when a dub would be wrong for the material
• Keep the whole localization pipeline on your own machine when the source media cannot be uploaded to a hosted service

// How YouDub WebUI compares

YouDub WebUI alongside other open-source audio, music & voice tools AI/TLDR tracks, ranked by GitHub stars.