What problem does it solve? Converting written scripts into spoken audio for voiceovers, narration, or podcasts normally requires recording equipment or manual API integration. This Skill provides a direct workflow for generating speech audio files from text through HeyGen's Starfish TTS model, with control over voice, speed, pitch, and locale. ## Core Features & Use Cases - Voice Discovery: List available Starfish-compatible TTS voices filtered by language, gender, and features via GET /v1/audio/voices or the mcp__heygen__list_audio_voices MCP tool. - Speech Generation: Convert text to audio with adjustable speed (0.5-1.5), pitch (-50 to 50), locale selection for multilingual voices, and optional ElevenLabs advanced settings. - Timing Data: Receive word-level timestamps in the response for caption syncing and timed text overlays, plus SSML break tags for inserting pauses. - Use Case: A content creator needs narration for a product demo video. They list voices, pick an English male voice, generate the audio with a 1.1x speed, and use the returned audio_url and word timestamps to sync captions. ## Quick Start Ask the AI to list available HeyGen TTS voices and generate speech audio from your script text using a selected voice.
