MCP.so
ログイン

Mcp Video2text

@yanyaoer

Mcp Video2text について

Convert online video to text, read it quickly

基本情報

カテゴリ

その他

ランタイム

python

トランスポート

stdio

公開者

yanyaoer

投稿者

yanyaoer

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "mcp-video2text": {
      "command": "uv",
      "args": [
        "run",
        "mcp-video2text"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Mcp Video2text?

Mcp Video2text is an MCP server that downloads online videos and transcribes their audio into text using Whisper models, enabling quick reading of video content. It is intended for users who want to extract text from publicly accessible online videos via any MCP client.

How to use Mcp Video2text?

Install the uv package manager and ffmpeg, then run uv pip install mcp-video2text[mlx] (recommended for Apple Silicon) or uv pip install mcp-video2text[cpp]. In your MCP client configuration, set the command to uv run mcp-video2text. The first execution downloads the Whisper model, which may take a few minutes; adjust your client’s timeout accordingly.

Key features of Mcp Video2text

  • Downloads online videos using yt-dlp
  • Transcribes audio to text with Whisper models
  • Supports MLX backend for Apple Silicon
  • Supports C++ backend (whisper.cpp)
  • Works with any MCP-compatible client

Use cases of Mcp Video2text

  • Transcribing educational or lecture videos for note-taking
  • Converting online talks or presentations to readable text
  • Extracting text from video tutorials for quick reference
  • Processing online videos where manual transcription is time‑consuming

FAQ from Mcp Video2text

What dependencies are required?

You need uv (package manager) and ffmpeg installed. On macOS, install ffmpeg via brew install ffmpeg; on Linux, use sudo apt install ffmpeg.

Does the first run require a long wait?

Yes, the first execution downloads the Whisper model, which can take several minutes. It is recommended to increase the timeout setting in your MCP client.

Where does the audio/video data live?

Data is downloaded and processed locally on your machine. The server uses yt-dlp to fetch the video, then runs Whisper locally for transcription.

Which transports or authentication methods are supported?

The server communicates over stdio transport as a standard MCP server. No authentication is mentioned in the README.

コメント

「その他」の他のコンテンツ