text-to-speech

To generate natural-sounding speech audio from text, you can use this Skill to call HeyGen /v1/audio endpoints, select a compatible voice, adjust speed and pitch, and produce narration without manual recording.

Yes, you can sync text-to-speech audio with captions by requesting optional word-level timestamps during speech synthesis, providing precise timing data needed for caption alignment in video edits.

Yes, you need a HEYGEN_API_KEY to authenticate and create text-to-speech jobs. The Skill uses this key to access HeyGen MCP tools or direct REST calls to the /v1/audio endpoints for audio generation.

The best way to list and choose HeyGen Starfish-compatible voices by language is to query available voices via MCP tools or /v1/audio REST endpoints, filtering by locale, gender, and preview audio before generating speech.

Yes, this text-to-speech approach supports multilingual voices and locale selection, allowing you to adjust the locale parameter for specific languages to produce localized narration audio.

You should use direct REST calls to the /v1/audio endpoints for speech synthesis as a fallback when HeyGen MCP tools are unavailable in your environment, ensuring text-to-speech job creation remains functional.