A local, high-accuracy speech-to-text tool powered by OpenAI Whisper for transcribing, translating, and summarizing audio in 100+ languages.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install voice-recognition
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 voice-recognition 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 Voice Recognition (Whisper)?
This skill provides a robust interface for OpenAI Whisper, enabling developers to perform speech-to-text processing entirely on their local machine. By leveraging Openclaw Skills, users can transcribe audio files across 100+ languages without relying on external API keys or incurring costs. It is ideal for privacy-conscious environments and workflows requiring high-fidelity transcription and automated summarization.
The system utilizes the medium Whisper model by default, striking an optimal balance between processing speed and transcription accuracy. Because it runs locally, it is a perfect addition to any developer's toolkit for handling sensitive audio data or building automated transcription pipelines.
Voice Recognition (Whisper) Use Cases
- Transcribing meeting notes or voice memos into text locally for improved documentation.
- Translating foreign language audio recordings directly into English for research and analysis.
- Generating concise summaries from long-form audio content to save time during review.
- Automating batch transcription workflows in high-privacy development environments using Openclaw Skills.
How Voice Recognition (Whisper) Works
- The user provides an audio file in a supported format like MP3, M4A, or WAV to the script.
- The script invokes the OpenAI Whisper CLI locally to process the audio data using the specified model.
- Optional flags are applied to specify the target language, trigger translation, or initiate summarization logic.
- The processed text is returned to the user, and the model remains cached locally for subsequent Openclaw Skills operations.
Voice Recognition (Whisper) Setup
First, ensure you have the necessary dependencies installed on your system:
# Install OpenAI Whisper CLI
brew install openai-whisper
# Ensure Python 3.10+ is installed
python3 --version
To simplify usage, you can add a quick command alias to your ~/.zshrc or ~/.bashrc file:
alias voice="python3 ~/.openclaw/workspace/scripts/voice识别_升级版.py"
Voice Recognition (Whisper) Data Schema & Taxonomy
The skill organizes its operations through the following components and data structures:
| Component | Description |
|---|---|
| Main Script | scripts/voice识别_升级版.py handles the logic and CLI arguments. |
| Model Storage | Models are downloaded and cached in ~/.cache/whisper to avoid repeated downloads. |
| Supported Formats | Compatible with MP3, M4A, WAV, OGG, FLAC, and WebM audio files. |
| Documentation | Internal technical details are stored in scripts/voice_tool_README.md. |
Voice Recognition (Whisper) Advanced Features
- Support for over 100 languages including Chinese (zh), English (en), Japanese (ja), and Korean (ko).
- Built-in translation flag (
--translate) to convert any supported language into English text automatically. - Summarization engine (
--summarize) to distill the core message from long transcriptions. - Local-first architecture ensures that your audio data never leaves your machine while using Openclaw Skills.
- Flexible model selection allows for balancing performance and accuracy based on hardware capabilities.
SKILL.md
Loading
Related Openclaw Skills
Attendance Sheet Generator
A specialized automation tool for generating professional, color-coded Excel attendance reports from employee work information.
Weather Assistant
A high-performance utility that fetches and summarizes real-time meteorological data for any specified city.
AI-NEWS
An automated intelligence gathering skill that synthesizes critical AI industry updates by sector and timeframe.
PaperReviser
An AI-driven platform that transforms academic reviewer feedback into actionable, structured revision suggestions for manuscripts.
RentaUnHumano MCP
A bridge between AI agents and real-world human tasks across Latin America via the Model Context Protocol.
AnkiConnect Skill
A specialized integration for managing Anki flashcards and decks using the AnkiConnect REST API within an AI agent environment.
Featured*
