An automated tool that generates speaker-attributed transcripts and subtitles from YouTube URLs with support for multiple file formats.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install youtube-transcript-speaker-diarization
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install youtube-transcript-speaker-diarization using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is YouTube Transcript Speaker Diarization?
The youtube-transcript-speaker-diarization skill is a specialized tool designed to produce structured, speaker-aware transcriptions from any public YouTube URL. Powered by the diarize.io service, this utility addresses the limitations of standard transcript tools by identifying individual speakers throughout the video timeline. This capability is useful for parsing multi-speaker discussions, interviews, podcasts, and detailed video reviews.
By integrating directly into your local workflows, this tool provides developer-focused automation via lightweight shell scripts. It delivers results in several target formats, including plain text, structured JSON, SRT, and VTT. It is built as a modular add-on for Openclaw Skills, allowing AI agents to seamlessly fetch, parse, and utilize structured dialog data from video files.
YouTube Transcript Speaker Diarization Use Cases
- Transcribing multi-speaker panel discussions, podcasts, or technical interviews from YouTube.
- Generating accurate SRT or VTT subtitles with proper speaker labels for video editing or accessibility workflows.
- Parsing structured dialogue data in JSON format for downstream NLP, LLM prompt injection, or database storage.
- Running long-running offline transcription tasks with check-in points using job status tracking.
How YouTube Transcript Speaker Diarization Works
- Submission: The user or AI agent triggers the shell script, providing a target YouTube URL. The script sends a request to the diarize.io API to initiate the transcription and diarization process.
- Job Queuing: The service creates a unique job ID representing the active transcription process.
- Polling & Status Monitoring: The script enters a polling loop to check the status of the transcription job on the server.
- Retrieval: Once the server completes the diarization, the script retrieves the transcription in the requested output format (TXT, JSON, SRT, or VTT) and returns it to the environment.
YouTube Transcript Speaker Diarization Setup
Prerequisites
Ensure you have curl and jq installed on your system.
# On macOS via Homebrew
brew install curl jq
Authentication
- Sign up for an account at diarize.io.
- Create an API key under your account settings.
- Export your API key as an environment variable:
export YOUTUBE_TRANSCRIPT_API_KEY="your_api_key_here"
Config Option
Alternatively, add the key directly to your configuration for Openclaw Skills:
{
"skills": {
"entries": {
"youtube-transcript-speaker-diarization": {
"apiKey": "YOUR_DIARIZE_API_KEY",
"baseUrl": "https://diarize.io"
}
}
}
}
YouTube Transcript Speaker Diarization Data Schema & Taxonomy
Output Formats
| Format | Description | Target Use Case |
|---|---|---|
txt |
Plain text with speaker tags | Human readability and rapid summaries |
json |
Structured JSON containing timestamps and speaker segments | LLM consumption and databases |
srt |
SubRip Subtitle file | Video players and basic captioning |
vtt |
Web Video Text Tracks format | Web-native video players and HTML5 |
Script Commands
The entrypoint script accepts the following commands:
run <url> [format]: Execute submission, polling, and output retrieval end-to-end.submit <url>: Submit a new job and return the jobId immediately.status <jobId>: Check the processing status of a submitted job.fetch <jobId> <format>: Download the finalized transcription in a specific format.
YouTube Transcript Speaker Diarization Advanced Features
- Asynchronous Execution Flow: Separate submit, status, and fetch phases to support long-running tasks without blocking agents or shell scripts.
- Format Flexibility: Simple parameters allow quick switching between subtitles (SRT, VTT), human-readable text (TXT), and raw machine-readable data (JSON).
- Robust Credentials Resolution: Automatically falls back across multiple environment variables and Openclaw Skills configuration schema definitions.
- Custom Base URLs: Option to override the default API endpoint for enterprise deployment, self-hosting alternatives, or testing proxy pathways.
SKILL.md
Loading
Related Openclaw Skills
Zhenghe Vault
Zhenghe Vault is a serverless, ERC-4626 yield vault on Base that allows AI agents to automatically grow USDC trading profits through network consumption rather than debt-based lending.
Meo Mai Moi MCP
A secure, OAuth-compliant Model Context Protocol gateway to interface AI assistants with the Meo Mai Moi pet care management system.
Foxtable Development Documentation
A split official-doc skill for Foxtable development that gives AI agents fast, accurate answers on tables, expressions, events, reports, and publishing.
Seewo Courseware Generator
Openclaw Skills generates native Seewo Whiteboard .enbx courseware from lesson topics, outlines, or raw teaching content, with validation before delivery.
Focus Toolkit
A zero-dependency productivity suite designed to manage focus sessions through natural language alarms, ambient sound generation, and detailed analytics.
Emoji Translator
A smart, zero-dependency translation engine that converts text into contextually relevant emojis and vice versa.
Featured*
