Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Reedapt MCP server Dub this video into Spanish: https://example.com/video.mp4"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Dub video and audio into other languages from any AI tool that speaks the Model Context Protocol: Claude Code, Claude Desktop, Cursor and others. Each speaker keeps their own voice, and the background sound is kept.
Make an API key under Developers in Reedapt. Test keys are free and dub the first 60 seconds of a video with a watermark; live keys come with paid plans and use plan minutes.
Claude Code
claude mcp add reedapt --env REEDAPT_API_KEY=rk_test_... -- npx -y reedapt-mcpClaude Desktop, Cursor and other clients (the mcpServers section of the client's config file):
{
"mcpServers": {
"reedapt": {
"command": "npx",
"args": ["-y", "reedapt-mcp"],
"env": { "REEDAPT_API_KEY": "rk_test_..." }
}
}
}Node.js 18 or later.
Related MCP server: OrcaDub MCP Server
Tool | What it does |
| The languages this key's plan can dub into |
| The plan, minutes left, or a month's usage report |
| Upload a local file or a public link and start one dub per language |
| A dub's status, stage and progress |
| Wait up to 4 minutes for a dub to finish; call again to keep waiting |
| The account's API dubs, newest first |
| Download links for a finished dub |
| The dub's lines, original and dubbed, 200 at a time |
| SRT or WebVTT, in the dub's language or the original's |
| Stop a dub that hasn't finished; its minutes are returned |
skill/reedapt/SKILL.md teaches an AI tool how to dub well with Reedapt: checking languages and minutes first, waiting, handing over the files, and what to say when something fails. To use it in Claude Code, copy the skill/reedapt folder into ~/.claude/skills/.
Stops a dub that hasn't finished. Its reserved minutes are returned. This can't be undone.
Inputs: dub_id (required).
Uploads a local file (an absolute path) or a public https link, and starts one dub per target language. Returns as soon as the dubs have started; dubbing takes minutes. Then call wait_for_dub or get_dub. Languages the plan can't dub into come back under refused with a reason. Live keys use plan minutes, so confirm the languages with the person first. Test keys dub only video files, and only the first 60 seconds.
Inputs: source (required), source_language (required), target_languages (required), metadata, +2 more.
A dub's status (queued, processing, completed or failed), its stage and progress, and the error if it failed.
Inputs: dub_id (required).
Download links for a completed dub: the dubbed video or audio, a WAV of the dubbed audio, and the separate voice and background tracks. Links expire at expires_at.
Inputs: dub_id (required).
A completed dub's subtitles as SRT or WebVTT text, in the dub's language or the original's.
Inputs: dub_id (required), format, language.
A completed dub's lines in time order: speaker, original words and the dubbed words, with times in milliseconds. Returns 200 lines at a time; use offset for more.
Inputs: dub_id (required), offset.
The account's plan, whether the key is live or test, and the minutes left. Check this before a long dub: a live dub uses one minute of the plan per minute of video, per language. Test keys are free and dub only the first seconds with a watermark.
Inputs: month.
The account's API dubs, newest first.
Inputs: limit, offset.
The language codes Reedapt can dub into right now, for this API key's plan.
Inputs: no named parameters.
Waits until a dub completes or fails, up to timeout_seconds (at most 240). If it is still running, it returns the latest status: call it again to keep waiting.
Inputs: dub_id (required), timeout_seconds.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 10 tools
Each tool has a clearly distinct role across the dubbing lifecycle: usage, starting, polling/waiting, listing, retrieving files/transcripts/subtitles, canceling, and languages. The closest pair, get_dub and wait_for_dub, is well differentiated by status-check versus blocking-wait behavior.
All tool names use consistent snake_case with predictable verb_noun structure (get_usage, dub_media, get_dub, list_dubs, cancel_dub). Even the less common wait_for_dub and dub_media fit the overall convention.
Ten tools is well-scoped for a media dubbing API, covering the full workflow without obvious bloat. Each tool corresponds to a meaningful operation rather than duplicating functionality.
The surface covers the full dubbing lifecycle: checking plan/usage, starting dubs, monitoring progress, waiting, listing dubs, downloading outputs, retrieving transcripts/subtitles, canceling, and listing available languages. No obvious operational gaps are apparent for the stated domain.
Transcribe, subtitle, dub and lip-sync media; create speech, voices, images and thumbnails.
Transcribe, subtitle and dub videos into 100+ languages, and translate text.
Translate your videos, check progress and retrieve dubbed results through your DubSync account.
Transcribe audio and video and work with your transcript library from any AI app. 119 languages.
