A robust offline speech-to-text skill utilizing whisper.cpp for high-speed, local audio transcription across multiple formats.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install whisper-transcriber
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 whisper-transcriber 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 Whisper Transcriber?
Whisper Transcriber is a professional-grade tool designed for local Automatic Speech Recognition (ASR) without relying on external cloud APIs. By leveraging whisper.cpp and ffmpeg, this skill provides a secure and efficient way to handle sensitive audio data. As part of the Openclaw Skills ecosystem, it focuses on performance and privacy, allowing users to generate high-accuracy transcriptions directly on their local hardware.
The skill is built to be cross-platform, offering seamless integration for macOS and Linux users, while providing a stable path for Windows users through WSL2. Whether you are transcribing a single voice note or processing an entire library of recordings, this implementation of Openclaw Skills ensures that your data remains under your control while benefiting from state-of-the-art AI models.
Whisper Transcriber Use Cases
- Transcribing meeting minutes and lectures into text or subtitles without uploading data to the cloud.
- Batch processing large directories of audio files for research or data archival.
- Generating SRT subtitle files for video production workflows locally.
- Automating voice-to-text pipelines within local development environments using Openclaw Skills.
How Whisper Transcriber Works
- The skill detects the input audio file or directory and verifies the existence of required binaries like ffmpeg and whisper-cli.
- It automatically checks for the specified GGML model in the local assets directory, downloading it if necessary.
- The audio is pre-processed and passed to the Whisper engine for high-speed inference.
- Transcription results are formatted into the user's choice of TXT, SRT, or JSON output.
- Temporary files are cleaned up automatically after the process completes to maintain a clean workspace.
Whisper Transcriber Setup
To get started with this skill, follow these installation and usage steps:
# Install dependencies and models
bash ./scripts/install.sh
# Transcribe a single audio file
bash ./scripts/transcribe.sh voice.ogg
# Batch transcribe a directory and output SRT subtitles
bash ./scripts/transcribe.sh ./recordings -b -s
Note: Windows users are highly recommended to run these Openclaw Skills within a WSL2 (Ubuntu) environment for optimal stability.
Whisper Transcriber Data Schema & Taxonomy
The skill organizes its resources and outputs using the following structure:
| Path | Description |
|---|---|
assets/models/ |
Default storage for GGML whisper models (ignored by git). |
scripts/ |
Contains the core transcription and installation logic. |
config.json |
Stores default configuration and metadata for Openclaw Skills. |
WHISPER_TEMP_DIR |
A configurable temporary directory for intermediate processing files. |
Whisper Transcriber Advanced Features
- Environment variable overrides for default models, languages, and custom model paths.
- Support for multiple output formats including SRT for video and JSON for programmatic analysis.
- Automated model management system that handles downloads and path resolution.
- Optimized for resource-constrained environments via the whisper.cpp implementation.
- Seamless batch processing mode for high-volume Openclaw Skills automation.
SKILL.md
Loading
Related Openclaw Skills
AIKEK API Crypto Research & Image Generation
A powerful suite of APIs for real-time cryptocurrency market analysis, on-chain research, and automated meme-style image generation.
VibeSurf Browser Automation
A comprehensive browser automation and web interaction framework that empowers AI agents to navigate the web, extract structured data, and manage external integrations.
PokerPal Game Assistant
A specialized AI agent skill for querying and managing live poker game data, including player statistics and financial settlements.
OpenClaw Model Router Skill
A robust model routing solution for OpenClaw featuring prefix-based triggers, timezone-aware scheduling, and fail-safe production switching.
Generate Daily Sports Update
A Python-based automation for crawling sports science research and wearable tech news with direct Notion synchronization.
Sports Science Daily Intelligence Engine
An automated sports science intelligence engine that aggregates 55+ global research and industry sources into localized, daily reports.
Featured*
