A comprehensive AI skill for building robust speech recognition, diarization, and realtime audio analysis integrations via SenseAudio.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install senseaudio-asr
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 senseaudio-asr 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 SenseAudio ASR?
SenseAudio ASR is a professional-grade integration designed for Openclaw Skills that enables developers to implement sophisticated speech-to-text functionalities. It supports offline file transcription, realtime WebSocket streaming, and detailed audio quality analysis, making it an essential tool for creating voice-enabled AI agents and applications.
By leveraging SenseAudio ASR, users can access specialized models tailored for different needs, including cost-effective transcription, intelligent editing with DeepThink, and complex multi-speaker diarization. This skill simplifies the process of handling audio data while maintaining high standards for security and performance in any AI-driven environment.
SenseAudio ASR Use Cases
- Realtime transcription of live microphone streams using WebSockets.
- Batch processing of offline audio files for transcription and translation.
- Multi-speaker diarization and sentiment analysis for meeting records.
- Audio quality pre-checks and noise analysis before performing recognition.
- Intelligent editing and streaming with the DeepThink model.
How SenseAudio ASR Works
- Select the appropriate recognition mode based on the audio source, such as HTTP for files or WebSocket for realtime streams.
- Choose a specific model like Lite, ASR, Pro, or DeepThink that aligns with your feature requirements such as diarization or translation.
- Construct a minimal request containing the required authentication header and audio format specifications.
- Validate that parameters and file sizes, typically kept under 10MB, meet the technical constraints of the API.
- Process the response robustly, handling various formats like JSON, Server-Sent Events (SSE), or WebSocket terminal events.
SenseAudio ASR Setup
To get started with this skill in your Openclaw Skills setup, configure your environment credentials:
# Set your SenseAudio API Key as an environment variable
export SENSEAUDIO_API_KEY="your_api_key_here"
Ensure that the API key is passed only via the Authorization: Bearer header and is never logged or included in query parameters.
SenseAudio ASR Data Schema & Taxonomy
| Component | Description | Supported Models |
|---|---|---|
| HTTP Transcription | Offline file processing for transcription and translation | Lite, ASR, Pro, DeepThink |
| WebSocket ASR | Realtime streaming recognition | ASR, DeepThink |
| Audio Analysis | Pre-recognition quality and noise checks | Dedicated Analysis Endpoint |
| Diarization | Speaker identification and separation | ASR, Pro |
| Metadata | Timestamps, sentiment, and trace IDs | ASR, Pro |
SenseAudio ASR Advanced Features
- Support for Server-Sent Events (SSE) to provide streaming text deltas for responsive interfaces.
- Intelligent editing and translation capabilities via the specialized DeepThink model.
- Explicit max_speakers control for high-precision diarization tasks in the Pro model.
- Realtime WebSocket integration supporting PCM 16000Hz mono audio requirements.
- Secure handling of sensitive operational data including trace_id and session_id for troubleshooting.
SKILL.md
Loading
Related Openclaw Skills
Email Assistant
An intelligent email management tool that automates categorization, professional drafting, and task extraction to streamline your daily communication workflow.
Atlassian Cloud APIs + CLIs
A unified interface for managing and automating the entire Atlassian Cloud ecosystem via first-party APIs and CLIs.
Chinese MOA Oxford-Style Debate
A sophisticated multi-agent system that simulates formal Oxford-style debates in Chinese to rigorously test arguments and explore multiple perspectives.
Multichannel Content Publisher
An AI-driven content adaptation engine that converts single messages into platform-specific publish packs with optimized formatting and CTAs.
Expedia
A professional-grade travel automation skill for searching Expedia inventory, validating real trip costs, and managing partner-safe booking workflows.
SenseAudio Voice Cloner
A specialized skill for guiding SenseAudio voice cloning and synthesizing personalized audio using cloned voice IDs.
Featured*
