A high-performance audio toolkit for decentralized text-to-speech, voice cloning, and transcription.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install deapi-audio
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 deapi-audio 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 deAPI Audio?
deAPI Audio is a powerful suite for audio manipulation and generation powered by a decentralized GPU network. It allows developers and creators to perform complex tasks like high-fidelity text-to-speech, instant voice cloning, and natural language voice design. By leveraging Openclaw Skills, this tool integrates directly into your AI agent's workflow, providing access to state-of-the-art models like Kokoro, Chatterbox, and Qwen3.
The skill is designed for versatility, handling everything from simple spoken text to professional-grade voice replication and multi-language transcription. Because it operates on a decentralized network, it offers a robust alternative to centralized providers, ensuring high availability and competitive performance for diverse audio processing needs.
deAPI Audio Use Cases
- Creating custom voiceovers for videos using natural language voice design.
- Replicating specific voices for personalized assistants through voice cloning.
- Transcribing podcasts or voice memos into text for documentation and search.
- Implementing multilingual text-to-speech for global application accessibility.
- Automating audio content creation via CLI scripts within AI agent workflows.
How deAPI Audio Works
- The user provides a text prompt or an audio file as input to the Openclaw Skills environment.
- The specific script (TTS, Clone, Design, or STT) processes the input parameters and validates the configuration.
- The request is dispatched to the deAPI decentralized GPU network for processing.
- The network executes the requested model (e.g., Qwen3 for design or Kokoro for fast TTS).
- The resulting audio file or transcription text is returned and saved locally or provided as a temporary URL.
deAPI Audio Setup
To get started with this skill, you need a deAPI API key and the necessary CLI utilities.
# Ensure curl and jq are installed
sudo apt install curl jq
# Configure your API key
echo '{ "api_key": "YOUR_DEAPI_API_KEY" }' > config.json
# Alternatively, export as an environment variable
export DEAPI_API_KEY="your_api_key_here"
deAPI Audio Data Schema & Taxonomy
The skill manages configuration and audio output through a structured file system within the Openclaw Skills directory.
| Data Type | Format | Location |
|---|---|---|
| Configuration | JSON | config.json |
| Audio Output | MP3/WAV | Script-defined output path |
| Scripts | Bash | /scripts/ directory |
| Metadata | YAML | SKILL.md frontmatter |
deAPI Audio Advanced Features
- Multi-model support allowing users to toggle between Kokoro, Chatterbox, and Qwen3 based on speed or quality needs.
- Advanced voice design that interprets descriptive text (e.g., "deep male voice with a British accent") to generate unique vocal profiles.
- High-accuracy voice cloning using reference audio samples and optional reference transcripts.
- Broad file format support for transcription including AAC, MP3, OGG, WAV, WebM, and FLAC.
- Automated language detection and support for over 22 languages across different model architectures.
SKILL.md
Loading
Related Openclaw Skills
Novel Writer Skill
An automated creative writing agent designed to draft high-quality long-form novels while maintaining strict narrative consistency and period-accurate details.
Deterministic Calc
A precision-focused skill that replaces AI numerical guesswork with reliable code-based execution for mathematical and shell operations.
Moark-TTS Gitee AI Suite
A comprehensive Text-to-Speech and voice feature extraction skill for Gitee AI, supporting over 15 state-of-the-art voice models.
Nano Banana 2
A professional-grade image generation and editing tool powered by Google's Gemini 3.1 Flash Image model for creating high-fidelity visual assets.
WeChat Publish Workflow
An end-to-end automation pipeline for WeChat Official Account publishing, covering everything from visual layout design to API-driven deployment and data analytics.
YouTube Content Manager Pro
An all-in-one AI-powered toolkit for managing the entire YouTube content lifecycle from topic discovery to performance analytics.
Featured*
