A privacy-focused offline speech-to-text skill utilizing the Vosk engine for local audio transcription.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install voice2text
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 voice2text 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 to Text?
The Voice to Text skill is a robust tool designed for high-performance, offline audio transcription. By leveraging the Vosk local model, this skill allows developers to process voice data into text without the need for an internet connection or external cloud APIs. As a specialized entry in the ecosystem of Openclaw Skills, it prioritizes data privacy and low-latency processing, making it suitable for secure environments and local automation tasks.
Voice to Text Use Cases
- Transcribing voice notes or audio recordings locally to maintain absolute data privacy.
- Automating transcription for media processing pipelines within the Openclaw Skills framework.
- Creating voice-controlled command-line interfaces that function without internet access.
- Extracting text from audio files for indexing or search engine optimization in offline databases.
How Voice to Text Works
- The skill receives a file path to a .wav audio file via the input parameters.
- It initializes the Vosk speech recognition engine using a locally stored model.
- The audio stream is processed through the model to identify speech patterns and phonemes.
- The system outputs a structured JSON object containing the final transcribed text.
Voice to Text Setup
To get started with this skill, install it via the hub and ensure your environment has the required speech recognition dependencies.
clawhub install voice2text
You can then run the skill by passing the path to your audio file:
clawhub run voice2text '{"audio":"sample.wav"}'
Voice to Text Data Schema & Taxonomy
The skill expects a simple input schema and returns a straightforward transcription object:
| Input Key | Type | Description |
|---|---|---|
audio |
string | The relative or absolute path to the .wav file. |
Output Format:
text: The string containing the resulting transcript from the processed audio.
Voice to Text Advanced Features
- Local inference using Vosk for high-speed processing without API costs.
- Easy integration with other Openclaw Skills for complex multi-agent workflows.
- Includes pytest support to ensure consistent transcription quality across different audio samples.
- Minimal resource footprint designed for CLI-based environments and edge devices.
SKILL.md
Loading
Related Openclaw Skills
SkillMixer
SkillMixer is an orchestration tool that automatically categorizes AI skills into logical groups and generates composite master skills.
Calm User Frustration
An emotional intelligence skill that detects user stress and intervenes with breathing exercises and automated wellness reminders.
Calm User Frustration
A wellness-oriented workflow for AI agents to detect user stress and provide immediate calming interventions and automated reminders.
Inbox Triage Bot
An AI-powered assistant that fetches, classifies, and summarizes your emails into actionable daily markdown reports.
GSTD Swarm Intelligence (gstd-a2a)
A decentralized protocol for autonomous agents to execute distributed tasks, perform multi-model consensus, and earn GSTD rewards on the TON blockchain.
gstd-a2a: Decentralized Agent-to-Agent Protocol
A decentralized protocol enabling autonomous agents to perform distributed tasks and earn cryptographic rewards on the TON blockchain.
Featured*
