Home/AI Glossary/Text-to-Speech (TTS)

Text-to-Speech (TTS)

Text-to-Speech (TTS) has evolved from robotic voices to neural voice profiles with pauses, emphasis, and sometimes real-speaker cloning. APIs make it easy to read responses from LLM out loud.

Always verify rights and consent for cloned voices. Tools: ElevenLabs, Descript, HeyGen. Counterpart: STT.

Key characteristics

  • Converts written text into synthetic speech with selectable voice, tone, and sometimes emotional style.
  • Is important for accessibility, voiceovers, voice assistants, and automated content publishing.
  • Requires consent checks, quality testing, and language support for reliable commercial use.