A sophisticated AI skill designed to convert raw voice recordings into professionally structured, high-impact meeting minutes and lecture summaries.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install speech-notes
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 speech-notes 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 Speech Notes?
The Speech Notes skill provides a comprehensive pipeline for turning audio into polished documentation within the Openclaw Skills ecosystem. It handles the entire lifecycle of speech processing, starting from technical audio optimization (segmentation and compression) to multi-engine transcription using high-performance models like Gemini and Qwen.
Beyond simple transcription, this skill excels at structural synthesis. It automatically organizes content into hierarchical headers, maintains the speaker's original tone while removing verbal filler, and applies advanced formatting logic to ensure the final output is ready for professional distribution. It is specifically optimized for collaborative environments like Feishu, utilizing block-level APIs for superior visual presentation.
Speech Notes Use Cases
- Transcribing and summarizing internal corporate meetings and strategic brainstorms.
- Converting academic lectures or guest speeches into structured, thematic study notes.
- Organizing disorganized personal voice memos into clear, actionable outlines.
- Processing existing raw transcripts to remove AI artifacts and improve professional readability.
How Speech Notes Works
- Analyze the input audio using ffprobe to determine duration and sample rates for optimal processing.
- Segment files longer than 5 minutes and compress them via ffmpeg to ensure high API success rates.
- Trigger the transcription engine (Feishu STT, Gemini, or Qwen-Omni) to produce a verbatim raw text.
- Apply a secondary LLM layer to restructure the text into a predefined Markdown hierarchy (H2 for themes, H3 for sub-topics).
- Refine the content by bolding key assertions, removing filler words, and inserting metadata into quote containers.
- Export the final structured document to local files or directly to external platforms via API.
Speech Notes Setup
To get started with this skill for Openclaw Skills, ensure you have the necessary CLI tools and environment variables configured:
# Install audio processing dependencies
brew install ffmpeg
# Set up the transcription script permissions
chmod +x scripts/speech-to-text.sh
# Configure your API credentials for your preferred provider
export FEISHU_FILE_KEY="your_key"
export GOOGLE_API_KEY="your_gemini_key"
export DASHSCOPE_API_KEY="your_qwen_key"
Speech Notes Data Schema & Taxonomy
The skill organizes data through a multi-stage lifecycle to ensure no loss of information while providing clean outputs:
| Component | Format | Description |
|---|---|---|
| Audio Segments | .mp3 / .wav | Compressed 5-minute chunks used for API ingestion. |
| Raw Transcription | .txt | Verbatim records stored locally for audit purposes. |
| Structured Doc | Markdown | Final summary with H2/H3 hierarchy and bolded logic points. |
| Metadata | JSON/Block | Contextual data including Speaker, Date, and Occasion. |
| Styling | Feishu API | Color-coded elements and background-highlighted quote blocks. |
Speech Notes Advanced Features
- Multi-model fallback logic: Automatically switches between Feishu STT, Gemini, and Qwen based on availability.
- Intelligent content 'Flavor' preservation: Retains speaker-specific terminology and first-person perspective while eliminating verbal clutter.
- Granular Block API Control: Direct integration with platform-specific APIs to modify text colors and container backgrounds without rewriting the entire document.
- Automated AI Anonymization: Removes all traces of AI generation to provide a document that looks manually edited by a professional secretary.
- Nesting Logic: Supports up to three levels of bullet point nesting for complex narrative chains.
SKILL.md
Loading
Related Openclaw Skills
OpenClaw Security Check
A fast, 10-point security auditor for OpenClaw configurations and host environments providing PASS/FAIL reports and automated remediation.
Gateway Watchdog
A proactive monitoring utility that detects abnormal error rates and spikes across multiple messaging channels to prevent service outages.
Feishu Document Readability Optimization
A technical skill for AI agents to maintain professional visual hierarchy and rich text integrity when programmatically editing Feishu or Lark documents.
Feishu Document Writing Assistant
An intelligent documentation skill that converts unstructured chat logs and meeting notes into professional, high-hierarchy Feishu documents.
Jenkins
A robust integration allowing AI agents to control and monitor Jenkins CI/CD servers directly through a REST API interface.
OpenClaw AVIF to JPG Converter
A powerful Python-based utility to batch convert AVIF images into high-quality JPEG format.
Featured*
