A specialized tool for transforming noisy Japanese speech-to-text transcripts into clean, structured smart home intents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install japanese-smart-home-command-normalizer
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 japanese-smart-home-command-normalizer 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 Japanese Smart Home Command Normalizer?
The Japanese Smart Home Command Normalizer is a robust utility designed to bridge the gap between imperfect voice recognition and reliable home automation. It specializes in correcting common Japanese STT drift, where phonetic errors like れーぼー (cooling) or だんぼー (heating) often lead to execution failures. By implementing this within your Openclaw Skills library, you ensure that messy transcripts are mapped to safe, structured slots.
Unlike broad LLM interpretations that can be unpredictable, this skill uses a combination of fixed vocabulary and lightweight fuzzy matching. This approach prioritizes safety and precision for home actions, ensuring that lighting and climate control commands are interpreted correctly even when the raw text input is degraded or informal.
Japanese Smart Home Command Normalizer Use Cases
- Normalizing noisy STT output from Japanese voice assistants before sending commands to IoT devices.
- Correcting frequent phonetic misinterpretations in Japanese smart home commands like aircon mode settings.
- Providing a safety layer that requests confirmation when voice command confidence is low.
- Standardizing device aliases such as 電気, ライト, and 照明 into a single light domain intent.
How Japanese Smart Home Command Normalizer Works
- The skill receives a short Japanese text string from an STT engine or audio-router hook.
- It consults the normalization pipeline defined in the design documentation to strip noise and handle fuzzy matching.
- The text is compared against the supported domains and vocabulary for devices like lights and air conditioners.
- It identifies the intended action (e.g., on, off, set_mode) and extracts required slots like temperature or mode.
- If the command is ambiguous or missing slots, it flags the result with a confirmation requirement.
- A structured JSON object is returned, ready for consumption by device-specific Openclaw Skills.
Japanese Smart Home Command Normalizer Setup
To integrate this normalization logic into your environment, follow these steps:
- Ensure you have the core logic available by referencing the library module:
# Navigate to your skill directory
cd skills/japanese-smart-home-command-normalizer
# Run the demo script to verify normalization logic
node scripts/demo.js
- Configure your existing device-control hooks (like SwitchBot) to pass raw transcripts through the
lib/normalize.jsmodule. - Review
references/openclaw-integration.mdto map the normalized output to your specific automation triggers.
Japanese Smart Home Command Normalizer Data Schema & Taxonomy
The skill organizes its logic and validation data through a structured file system:
| Component | Description |
|---|---|
lib/normalize.js |
The core JavaScript module performing the text-to-intent transformation. |
fixtures/samples.json |
A collection of test cases containing raw transcripts and expected normalized outcomes. |
references/domains.md |
The source of truth for supported vocabulary, aliases, and device modes. |
references/design.md |
Documentation for the API response shape and the confidence-scoring logic. |
Normalized results typically include domain, action, slots, and a needsConfirmation boolean flag.
Japanese Smart Home Command Normalizer Advanced Features
- Fuzzy matching for common Japanese STT errors (e.g., エアコントメテ to aircon stop).
- Multi-domain support covering both lighting and HVAC systems with extensible vocabulary.
- Safety-first confirmation logic that prevents accidental execution on low-confidence transcripts.
- Seamless integration with the Openclaw Skills ecosystem via standardized thin-hook guidance.
SKILL.md
Loading
Related Openclaw Skills
Mac Health Check
A specialized skill for monitoring real-time thermal, power, and performance telemetry on Apple Silicon Macs.
Agent Memory Journal
A sophisticated 3-tier memory management system for AI agents to capture, promote, and retrieve long-term information.
Village Carpenter Starter
A practical framework for AI agents to guide users through establishing a manual woodworking trade using scavenged materials and essential hand tools.
Village Blacksmith Starter
A comprehensive AI agent skill that guides users through building a coal forge from scrap and establishing a self-sustaining blacksmithing trade.
OkraPDF Public Documents
A specialized MCP skill for AI agents to instantly query, read, and extract structured data from pre-indexed Arxiv papers and SEC filings.
Aruba CX Switch Security Hardening Tool
A professional-grade API tool for generating CIS Benchmark-compliant security configurations for Aruba CX network switches.
Featured*
