Text-to-Speech

This skill converts written text into natural-sounding speech using ElevenLabs' voice synthesis technology. It supports over 70 languages with six different models optimized for different use cases—from highest-quality emotional speech (eleven_v3) to ultra-low latency real-time applications (eleven_flash_v2_5). You can select from pre-built voices or create custom ones, and fine-tune output with stability, similarity boost, style, and speaker boost settings. The skill also offers request stitching to eliminate audio artifacts across long-form content, language enforcement for pronunciation control, text normalization for natural number and date reading, and multiple output formats including MP3, PCM, Opus, and telephony-standard μ-law. Developers building voice applications, voiceover systems, multilingual content platforms, or real-time conversational AI benefit from its flexible architecture and streaming capabilities.

Install the skill using the command provided and refer to the setup guide for additional details.