A high-performance, local text-to-speech engine optimized for Spanish voice synthesis and mobile messaging.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install vibevoice
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 vibevoice 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 VibeVoice TTS?
VibeVoice is a sophisticated local text-to-speech solution that leverages Microsoft's VibeVoice model to produce lifelike audio. It is designed specifically to integrate within Openclaw Skills to provide private, GPU-accelerated voice generation without relying on external cloud APIs.
This skill excels at producing Spanish speech with natural prosody, making it ideal for automated communication systems. By running locally, it ensures data privacy and eliminates latency issues associated with network-dependent services while maintaining high-quality output compatible with various media formats.
VibeVoice TTS Use Cases
- Sending natural-sounding WhatsApp voice messages via automated agents.
- Generating local audio feedback for AI-driven desktop interactions.
- Creating Spanish language voiceovers for content using Openclaw Skills.
- Developing offline-capable speech interfaces for secure environments.
How VibeVoice TTS Works
- The user or agent triggers the generation script with text input and optional parameters like voice or speed.
- The system initializes the VibeVoice model through a local Python environment established for Openclaw Skills.
- Input text is processed by the GPU-accelerated model to generate high-fidelity raw audio data.
- ffmpeg automatically converts the output into the desired format, such as .ogg for mobile compatibility.
- The generated file is saved to the specified directory for immediate use or transmission via messaging tools.
VibeVoice TTS Setup
To get started with this skill, follow these installation steps:
git clone https://github.com/microsoft/VibeVoice.git ~/VibeVoice
cd ~/VibeVoice
python3 -m venv venv
source venv/bin/activate
pip install -e .
pip install torch torchaudio
Ensure that ffmpeg and an NVIDIA GPU with at least 2GB VRAM are available to support the performance requirements of Openclaw Skills.
VibeVoice TTS Data Schema & Taxonomy
The skill manages audio outputs and model configurations using the following structure:
| Component | Details |
|---|---|
| Output Formats | Supported: .ogg (Opus), .mp3, .wav |
| Default Voice | sp-Spk1_man (Spanish male with Mexican accent) |
| Configuration | Adjustable speed (0.5 - 2.0) and custom voice profiles |
| Storage | Local model storage in ~/VibeVoice/demo/voices/ |
VibeVoice TTS Advanced Features
- Native support for Opus codec (.ogg) ensuring seamless WhatsApp voice message integration via Openclaw Skills.
- Real-time performance with a 0.24x Real-Time Factor (RTF), generating one minute of audio in approximately 15 seconds.
- Multi-language capabilities including specific high-quality English and Spanish speaker profiles.
- Direct CLI integration for flexible automation within sophisticated AI agent workflows.
SKILL.md
Loading
Related Openclaw Skills
Sound FX (ElevenLabs)
A high-performance utility for generating custom sound effects from text prompts using the ElevenLabs API.
A read-only Reddit scraper for AI agents that allows searching and browsing subreddits via the public JSON API without an API key.
DingTalk Calendar Management
An AI-powered skill to manage DingTalk calendars, attendee status, and meeting room reservations via the mcporter CLI.
i18n for Next.js
A comprehensive implementation guide and toolset for full internationalization (i18n) and localized SEO in Next.js App Router applications.
Keyswap
A specialized utility designed to automate the rotation and update of Claude Max API tokens across multiple OpenClaw Anthropic profiles.
EmailAgent
An intelligent email assistant that uses LangChain and OpenAI to draft and send emails with a mandatory human approval step.
Featured*
