qwen3-tts

Generate multilingual speech with voice cloning and Telegram/WhatsApp delivery.

Updated Feb 20, 2026

Search Tags:#whatsapp #tts #telegram #openai-api #voice-design #voice-clone #gpu-detection

Ready to run

npx skills add https://github.com/daMustermann/claw-qwen3-tts --skill qwen3-tts-damustermann

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: qwen3-tts
Source: https://github.com/daMustermann/claw-qwen3-tts/tree/main
Command: npx skills add https://github.com/daMustermann/claw-qwen3-tts --skill qwen3-tts-damustermann

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

OpenClaw agents often need reliable, multilingual text-to-speech with voice customization and easy distribution of generated audio. This skill provides a turnkey TTS system with built-in voices, voice design and cloning capabilities, and native delivery of audio via Telegram/WhatsApp.

Core Features & Use Cases

10 built-in speakers for rapid voice selection and storytelling.
Voice design from natural language descriptions to create new personas.
Voice cloning from as little as 3 seconds of reference audio, with persistent voice saving by name.
OpenAI-compatible API endpoints for speech, voices management, and format conversion.
Telegram and WhatsApp integration to deliver audio as native voice messages.

Quick Start

Install the skill, start the server, and generate your first speech sample via the v1 endpoint.

Frequently Asked Questions about qwen3-tts

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I clone a voice from a short audio sample for text-to-speech?▼

Voice cloning for text-to-speech requires as little as 3 seconds of reference audio to create a new voice. The system saves these cloned voices persistently by name, allowing you to reuse them for generating speech in your agents.

Can I deliver generated text-to-speech audio as native voice messages on Telegram and WhatsApp?▼

Yes, you can deliver generated text-to-speech audio as native voice messages directly through Telegram and WhatsApp. The skill handles end-to-end audio conversion and message delivery natively for both platforms.

Does this text-to-speech skill provide an OpenAI-compatible API?▼

Yes, this text-to-speech skill provides a local OpenAI-compatible API endpoint. It supports speech generation, voice management, and audio format conversion directly through this interface.

How can I design a custom TTS voice using natural language descriptions?▼

You can design a custom TTS voice by providing a natural language description of the desired persona. The voice design feature generates high-quality, multilingual speech from this text prompt.

Do I need a GPU to run this multilingual text-to-speech server?▼

The skill includes built-in GPU detection to optimize the text-to-speech model loading and generation process. While it handles the local server setup automatically, having a GPU ensures faster processing for high-quality speech synthesis.

What is the best way to manage persistent named voices for OpenAI-compatible text-to-speech?▼

The best way to manage persistent named voices is to use the built-in voice management API. You can save both cloned and designed voices by name, ensuring they remain available for reuse across multiple text-to-speech generations.