Supertone TTS

Wraps the Supertone TTS API so you can synthesize speech, clone voices, and browse a multilingual catalog directly from Claude or any MCP client. Exposes text_to_speech with pitch shift and speed controls, predict_duration for cost estimation before synthesis, search_voice to filter by language/gender/age/use case, and clone_voice to create custom voices from local audio files. Supports Korean, English, Japanese, and 20+ other languages. Output modes let you save files to disk, return inline audio resources, or both. Includes full CRUD for managing cloned voices. Useful when you need programmatic TTS without leaving your LLM workflow or want to test voice samples before committing credits.