clawvox

What This Skill Does

ClawVox transforms your OpenClaw assistant into a professional-grade voice production studio powered by ElevenLabs. This skill serves as a comprehensive interface for voice-based AI tasks, allowing you to generate high-quality speech from text, transcribe audio files into text, clone existing voices for personalized narration, and even generate sound effects for multimedia projects. With support for advanced features like multilingual generation, voice isolation for removing background noise, and automated dubbing, ClawVox is the central hub for all audio-related agent workflows within the OpenClaw ecosystem.

Installation

To install the ClawVox skill, run the following command in your terminal: clawhub install openclaw/skills/skills/abhishek-official1/clawvox. Once installed, you must authenticate the skill by providing your ElevenLabs API key. You can add this directly to your configuration file at ~/.openclaw/openclaw.json under the skills.entries.clawvox section or by setting the ELEVENLABS_API_KEY environment variable in your shell profile. Proper configuration ensures the agent can communicate with the ElevenLabs cloud services to process your audio requests.

Use Cases

ClawVox is highly versatile and fits into several professional and creative workflows. For content creators, it simplifies the production of high-quality voiceovers for videos, podcasts, or audiobooks using realistic AI models like eleven_multilingual_v2. Developers can utilize the transcription and voice isolation tools to build accessible applications that process user audio input. For global teams, the dubbing feature provides an efficient way to translate spoken content into other languages while preserving the speaker's original vocal characteristics. Additionally, the ability to generate sound effects from simple text descriptions makes it an essential tool for game developers and audio designers.

Example Prompts

1. "ClawVox, generate a professional voiceover for my new product presentation using the 'Adam' voice at a normal speaking speed."
2. "Please transcribe the meeting audio file located at ~/recordings/interview.mp3 and save the text output to my desktop."
3. "Create a stormy wind sound effect for the background of my game scene and save it to the audio library."

Tips & Limitations

To get the best results, use the eleven_multilingual_v2 model for professional narration, but prefer eleven_flash_v2_5 for real-time interactive responses to keep latency low. Keep in mind that cloning a voice requires a high-quality, clean sample to ensure accuracy; avoid background music or echo in your source audio. Always verify your remaining API usage quota via your ElevenLabs account dashboard, as high-volume audio generation and long-form transcription can consume your credits quickly. If you experience poor transcription quality, use the isolate.sh script first to clean up the input audio before running the transcription task.