elevenlabs

The ElevenLabs clinical TTS skill is a specialized utility designed for healthcare developers and OpenClaw users to transform written medical documentation into high-quality, natural-sounding audio. It is built to improve patient accessibility and engagement by providing empathetic voice output for sensitive healthcare scenarios. The skill leverages ElevenLabs' advanced text-to-speech models, including multilingual support for over 30 languages, making it an essential tool for creating inclusive clinical communications, such as discharge instructions, medication schedules, and post-operative care reminders.

• Employs eleven_multilingual_v2 for high-fidelity language synthesis, ideal for patient-facing instructions where clarity and empathy are paramount.

• Supports low-latency performance via eleven_flash_v2_5 for real-time reminders and immediate interaction scenarios.

• Enables the creation of accessible voice content from clinical text, helping patients with visual impairments or reading difficulties better understand their health data.

• Facilitates multilingual health messaging, ensuring that non-English speakers receive clear, professionally toned medical guidance in their preferred language.

• Provides configurable parameters for voice selection, enabling users to choose calm, authoritative, and professional tones suitable for clinical environments.

• Ensure all API usage adheres to security and privacy guidelines; never include Protected Health Information (PHI) or personally identifiable data in log files or API payload logs.

• Use short, clear sentence structures for medical instructions to optimize TTS pacing and listener comprehension.

• Utilize pronunciation hints or dictionaries if the specific clinical terminology is mispronounced by the default model.

• The tool expects a valid ELEVENLABS_API_KEY as an environment variable or configuration file entry at ~/.openclaw/openclaw.json.

• Typical outputs are audio streams formatted for immediate playback in patient portals, telehealth platforms, or local OpenClaw voice agents.