Pocket-TTS is a lightweight, CPU-optimized text-to-speech engine that enables real-time streaming and voice cloning without requiring a GPU.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install lb-pocket-tts-skill
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 lb-pocket-tts-skill 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 Pocket-TTS?
Pocket-TTS is a high-performance audio synthesis tool designed to run efficiently on local hardware. As a prominent addition to Openclaw Skills, it leverages a 100M parameter model to deliver high-quality English speech generation. Unlike many modern AI models that demand heavy GPU resources, this skill is specifically tuned for CPU efficiency, utilizing only two cores to achieve speeds up to 6x faster than real-time on modern processors like the M4.
This skill is ideal for developers who need to integrate low-latency voice responses into their applications while maintaining data privacy. By supporting streaming generation with a latency of approximately 200ms to the first chunk, it provides a seamless interactive experience. Whether you are building an automated assistant or local media tools, this entry in the Openclaw Skills library ensures professional results without external API dependencies.
Pocket-TTS Use Cases
- Generating local speech from text for privacy-focused AI applications.
- Creating custom voice clones from short 3-10 second audio samples.
- Implementing low-latency streaming for real-time conversational agents.
- Batch processing text files into 24kHz mono WAV audio for content creation.
- Running a local, self-hosted TTS server with a built-in web interface.
How Pocket-TTS Works
- The user installs the package and loads the 100M parameter lightweight model into memory.
- A voice state is established by selecting a default voice or providing an audio sample (3-10s) for instant cloning.
- The engine processes the input text through the CPU-optimized model using specific quality tuning parameters like temperature and LSD decode steps.
- Audio is synthesized into high-quality 24kHz mono WAV format.
- The output is either saved as a file, streamed in chunks for immediate playback, or served via a FastAPI web server.
Pocket-TTS Setup
To begin using this skill from the Openclaw Skills collection, install it via your preferred Python package manager:
# Standard installation via pip
pip install pocket-tts
# Or using uv for faster management
uv add pocket-tts
Once installed, you can generate your first audio file directly from the command line:
pocket-tts generate --text "Welcome to the Openclaw Skills ecosystem."
Pocket-TTS Data Schema & Taxonomy
Pocket-TTS organizes its outputs and assets with a focus on speed and compatibility. Below is the metadata and file structure usage:
| Data Type | Description | Format |
|---|---|---|
| Audio Output | High-quality mono speech | .wav (24kHz, 16-bit PCM) |
| Voice State | Pre-processed voice embeddings for fast loading | .safetensors |
| Input Prompts | Source audio used for voice cloning | .wav, .mp3, or URL |
| Model Config | Parameters for temperature and decode steps | CLI Arguments / JSON |
Pocket-TTS Advanced Features
- Rapid voice cloning that transforms 3-10 seconds of source audio into a reusable voice state.
- Direct streaming API allowing for real-time audio playback with sub-200ms latency.
- Support for .safetensors export to enable near-instant loading of custom voices.
- Built-in FastAPI integration for deploying a text-to-speech web service on local networks.
- Multi-voice management capabilities within the Python API for dynamic persona switching.
SKILL.md
Loading
Related Openclaw Skills
Motion.dev Animation Library
A high-performance animation skill for implementing Motion.dev (formerly Framer Motion) across React, Vue, and vanilla JavaScript projects.
BMad Method
A 4-phase agile framework for AI coding agents to manage the full software development lifecycle from analysis to implementation.
Better Auth Documentation Integration
A comprehensive technical reference skill for implementing modern, secure authentication in TypeScript projects using AI agents.
Xiangji Translation Service
Xiangji Translation Service provides professional-grade text and image translation capabilities by integrating multiple global AI engines.
shadcn/ui Documentation Skill
A comprehensive, AI-ready documentation package for shadcn/ui components, theming, and framework integrations.
Tailwind CSS Documentation
A comprehensive documentation skill providing AI agents with instant access to Tailwind CSS utility classes, configuration patterns, and best practices.
Featured*
