YouTube Dubbing MCP
MCP server that dubs YouTube videos or SRT files, with one cloned voice per speaker
- Pricing
- Open source
- Platforms
- CLI, self-hosted, macOS
- Runs
- Locally
- Version
- 0.1.0
- Listing updated
- 3 days ago
About YouTube Dubbing MCP
YouTube Dubbing MCP is an open-source MCP server for generative video. It runs locally on your machine and is started by your AI client.
The agent does the translating and this server handles everything that touches audio. It pulls the transcript, works out who is speaking from caption markers or voice pitch, cuts a sample of each speaker from the source and clones that voice, then renders, times and mixes the dub. A project folder keeps progress, so re-running a tool reuses finished lines and never submits the same line twice.
Timing is handled with a character budget per line for the target language; lines start with the original timing and are only sped up, capped at 1.25 times, when needed, so the track matches the video length that YouTube requires. Output is organised per language with an m4a audio file, an SRT, a localized title and description and the YouTube Studio upload steps. One of the tools the README names is dubbing_extract_voice_samples.
It runs from a Docker image or via uvx, with ffmpeg bundled and no local GPU. Voices are rendered by the Miragea voice engine, which needs an API key. The README calls it open source.
Install
Run in a terminal
claude mcp add youtube-dubbing-mcp --env MIRAGEA_API_KEY='<MIRAGEA_API_KEY>' -- docker run -i --rm -e MIRAGEA_API_KEY docker.io/miragea/youtube-dubbing-mcp:0.1.0Add to claude_desktop_config.json
{
"mcpServers": {
"youtube-dubbing-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"MIRAGEA_API_KEY",
"docker.io/miragea/youtube-dubbing-mcp:0.1.0"
],
"env": {
"MIRAGEA_API_KEY": "<MIRAGEA_API_KEY>"
}
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"youtube-dubbing-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"MIRAGEA_API_KEY",
"docker.io/miragea/youtube-dubbing-mcp:0.1.0"
],
"env": {
"MIRAGEA_API_KEY": "<MIRAGEA_API_KEY>"
}
}
}
}Add to .vscode/mcp.json
{
"servers": {
"youtube-dubbing-mcp": {
"type": "stdio",
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"MIRAGEA_API_KEY",
"docker.io/miragea/youtube-dubbing-mcp:0.1.0"
],
"env": {
"MIRAGEA_API_KEY": "<MIRAGEA_API_KEY>"
}
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"youtube-dubbing-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"MIRAGEA_API_KEY",
"docker.io/miragea/youtube-dubbing-mcp:0.1.0"
],
"env": {
"MIRAGEA_API_KEY": "<MIRAGEA_API_KEY>"
}
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"youtube-dubbing-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"MIRAGEA_API_KEY",
"docker.io/miragea/youtube-dubbing-mcp:0.1.0"
],
"env": {
"MIRAGEA_API_KEY": "<MIRAGEA_API_KEY>"
}
}
}
}Configuration
MIRAGEA_API_KEYVoice engine API key from https://miragea.ai/workspace/apiRequired
Tools this server provides (1)
Taken from the project documentation; the running server may differ.
- dubbing_extract_voice_samples
- Cuts a clean voice sample of each speaker from the video.
What it does
- Voice per speaker
- Samples are cut from the video itself so each speaker keeps their own cloned voice.
- YouTube or local input
- Takes a YouTube URL, local MP4, MOV, MKV, MP3 or WAV files, or SRT and VTT subtitles.
- Timing control
- Per-line character budgets and capped speed-up keep the dub as long as the video.
- Upload-ready folders
- Each language gets audio, captions, a localized title and description.
- Resumable projects
- Finished lines are reused when a tool is re-run.
Good for
- Dub an interview into English while keeping both speakers' voices
- Produce multi-language audio tracks for a YouTube channel
- Turn an SRT file into a spoken dub
- Create localized titles and descriptions alongside audio
Pricing
Open source
Server is open source; voice rendering needs a Miragea API key, prices not published here
Prices change. Check huggingface.co before you buy.
YouTube Dubbing MCP FAQ
What API key is needed?
A Miragea voice engine key, set as MIRAGEA_API_KEY.
Does it need a GPU?
No. The README says there are no local models and ffmpeg is bundled.
Who translates the text?
The connected agent's own model translates line by line within the time budget.
How is it installed?
As a Docker image or with a single uvx command.
Is YouTube Dubbing MCP free?
Yes. YouTube Dubbing MCP is open source, so you can use and self-host it for free.
How do I add YouTube Dubbing MCP to Claude Code?
Run this command in your terminal: claude mcp add youtube-dubbing-mcp --env MIRAGEA_API_KEY='<MIRAGEA_API_KEY>' -- docker run -i --rm -e MIRAGEA_API_KEY docker.io/miragea/youtube-dubbing-mcp:0.1.0
What are the best alternatives to YouTube Dubbing MCP?
The closest alternatives to YouTube Dubbing MCP are HyperFrames by HeyGen, ffmpeg-skill, AvatarSay — Fictional Character & Talking Video MCP, ChartAnt and VfxSmith.
Alternatives to YouTube Dubbing MCP

HyperFrames by HeyGen
MCP serverHosted MCP connector for creating and rendering videos from HTML through AI chat
- Remote
- Freemium
- 85.3K GitHub stars
- Design and media

ffmpeg-skill
MCP serverLocal FFmpeg video and audio editing tools for coding agents over MCP
Runs locally- Open source
- 6.9K GitHub stars
- Design and media

Generate fictional character portraits and lip-synced talking videos from text over MCP
- 10 tools
- Remote
- Freemium
- Design and media

ChartAnt
MCP serverRemote MCP server returning charts, maps and dashboards as hosted PNG links
- 51 tools
- Remote
- Freemium
- Design and media

VfxSmith
MCP serverSearch, generate, refine and export pixel art game effects from Claude Code or Codex
Not checked yet- Remote
- Freemium
- Design and media
Eikino
MCP serverEstimate where photos or video frames were taken and check media for AI generation
- 3 tools
- Remote
- Freemium
- Security