Efficiently convert audio and video files into high-accuracy text and structured summaries using professional-grade ASR models.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install audio-transcribe-summarize
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 audio-transcribe-summarize 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 Audio Transcribe & Summarize?
The Audio Transcribe & Summarize skill is a powerful extension for Openclaw Skills designed to bridge the gap between spoken content and actionable text. By leveraging the SenseAudio ASR API, this skill automates the process of transcribing audio or video files and synthesizing them into structured formats like meeting minutes, lecture notes, or interview highlights. It is particularly valuable for developers and teams who need to process large volumes of multimedia data without manual effort.
This tool handles everything from basic transcription to advanced diarization and sentiment analysis. With built-in support for splitting large files and multiple ASR model tiers—including high-accuracy and deep-thinking variants—it ensures that every recording is processed with optimal precision and context-awareness.
Audio Transcribe & Summarize Use Cases
- Automated meeting minutes generation with speaker identification and action items.
- Converting academic lectures or podcasts into searchable, structured notes for personal knowledge management.
- High-accuracy transcription of legal or medical interviews using deep-thinking ASR models.
- Translating foreign language audio into English text for global content analysis.
- Processing high volumes of customer service recordings for sentiment analysis and key insight extraction.
How Audio Transcribe & Summarize Works
- The user provides an audio or video file in a supported format like MP3, WAV, or MP4.
- The skill assesses the file size and uses ffmpeg to automatically split files larger than 10MB to comply with API limits.
- A specific ASR model (lite, standard, pro, or deepthink) is selected based on the user's accuracy and cost requirements.
- The transcription script communicates with the SenseAudio API to perform ASR, speaker diarization, and sentiment analysis.
- The resulting transcript is processed to generate a structured summary including key points, action items, and notable quotes.
Audio Transcribe & Summarize Setup
To integrate this capability into your Openclaw Skills workflow, follow these steps:
- Obtain a SenseAudio API key from the SenseAudio platform.
- Configure your environment variable:
export SENSEAUDIO_API_KEY='your_api_key_here'
- Ensure Python 3.8+ and necessary dependencies are installed:
pip install requests
- For large file support, install ffmpeg:
# macOS
brew install ffmpeg
# Ubuntu/Debian
sudo apt install ffmpeg
Audio Transcribe & Summarize Data Schema & Taxonomy
This skill organizes output data alongside the source files for easy access. The following schema defines the file taxonomy:
| Data Type | Description |
|---|---|
| Source File | Original audio/video (wav, mp3, mp4, etc.) |
| Transcript (.txt) | Full text output generated by the ASR model |
| Structured Summary | Markdown file containing key points, action items, and quotes |
| Metadata | Speaker labels, timestamps, and sentiment scores per segment |
The skill uses scripts within its directory and generates transcription outputs in the same location as the source media.
Audio Transcribe & Summarize Advanced Features
- Multi-speaker diarization for accurately tracking conversations in group settings.
- Integrated sentiment analysis to detect emotional tone throughout the recording.
- Automatic translation capabilities for converting transcripts directly into English.
- Flexible ASR model selection including deepthink for noisy environments or jargon-heavy audio.
- Collapsible transcript views in the final summary for better readability of long-form content.
SKILL.md
Loading
Related Openclaw Skills
Fluid Network Analyzer
A professional industrial skill for modeling, solving, and analyzing fluid networks using TOML-based definitions and Markdown reporting.
PV_15 Chief Risk Officer
An authoritative AI risk management agent designed to provide cold, analytical oversight and veto power for complex decision-making workflows.
GitHub CLI Integration
A powerful integration allowing AI agents to manage repositories, pull requests, and CI/CD workflows using the GitHub CLI.
Weather Agent Skill
A zero-configuration utility to fetch real-time weather data and multi-day forecasts directly via CLI using free, high-availability services.
PV_13: Venture Capitalist AI Agent
An assertive, goal-oriented AI persona built to drive exponential growth and disruptive innovation through strategic resource allocation.
Web to PDF Converter
A specialized skill to transform live web pages into high-fidelity PDF documents using Playwright.
Featured*
