“Get the transcript of this YouTube talk and list the audience questions with their times.”
- app_run · video-to-text
- app_run_status
- app_run_result
The transcript, plus each audience question with the time of the segment it's in.
Docsie MCP server · Video to text
Get a video's transcript in Claude, ChatGPT or your IDE.
Give your assistant a YouTube, Loom, Vimeo or Google Drive link, a direct video URL or a video in your Docsie workspace. The Docsie Video to Text app returns the transcript text plus timed segments, so your assistant can quote, summarize or search it right in the chat.
Runs the same Video to Text app as Docsie. Uses AI credits.
https://app.docsie.io/mcpWorks in Claude, ChatGPT, Claude Code, Cursor, Codex and any MCP client.
The Video to Text app
In Docsie you upload a video, paste a link or pick one from Google Drive. The same app runs over MCP, and the transcript comes back to your assistant.
What you can ask
Ask for the words, with their times.
Your assistant runs the Video to Text app and reads back the transcript. Each segment carries its time, so answers can point to the moment in the video.
“Transcribe this Loom update and summarize the decisions in five bullets.”
- app_run · video-to-text
- app_run_result
A summary written from the transcript text, with the transcript there to check it against.
“Pull the transcript of our recorded design review and find every mention of the payments API.”
- app_run · video-to-text
- app_run_result
Each mention with its timed segment, so you know where to look in the recording.
Watch it work
See a video turned into a timestamped transcript.
Walkthroughs recorded in the Docsie app. Over MCP, your assistant runs the same app and gets the same results back in the chat.
How it works
How video to text works over MCP.
Share the video
Paste a YouTube, Loom, Vimeo or Google Drive link or a direct video URL, or point to a video already in your Docsie workspace.
Run Video to Text
Your assistant starts the app with app_run, checks on it and reads the result when it's ready.
Use the transcript
You get the text plus timed segments. Ask for quotes, summaries or notes, each tied to its time.
Permissions and limits
What to tick, and what it costs.
Tick Workspace create and edit
Running an app needs the Workspace create and edit pack, which is off by default. Reading results only needs Workspace search and read.
AI credits
Every run uses your organization's AI credits, the same as running Video to Text in Docsie. The MCP server itself is on every plan.
Transcript and timed segments
Video to Text returns the words and their times. For a written guide or SOP from the same video, ask for video to docs instead.
About Docsie Video to Text →FAQ
Common questions
Can Claude get a YouTube transcript?
Yes, once it's connected to the Docsie MCP server. Paste the link and ask for the transcript. Claude runs the Video to Text app and gets the text plus timed segments back.
Is there a YouTube transcript MCP server for ChatGPT?
The Docsie MCP server works in ChatGPT's developer mode. ChatGPT can run Video to Text on a YouTube link and read the transcript in the chat. Each person signs in with their own Docsie account.
Which videos work?
YouTube, Loom, Vimeo and Google Drive links, direct video file URLs, and videos already in your Docsie workspace.
What are the timed segments for?
Each piece of the transcript comes with its time, so your assistant can tell you where a quote or topic appears in the video.
Does it cost extra?
The MCP server is on every Docsie plan, Free included. Each transcript run uses AI credits, the same as in the app.
Docsie MCP server
Get transcripts from the tools you already use.
Connect Claude, ChatGPT or Cursor to Docsie and pull a video's transcript into the chat.