Definition
Text-to-speech (TTS), also called speech synthesis, is the technology that converts written text into audible speech output.
How it works
Classic pipelines normalize the text first (numbers, abbreviations, pronunciations), generate speech features with an acoustic model, and reconstruct the waveform with a vocoder; newer neural TTS systems are largely end-to-end and can clone a voice from a short audio sample.
Usage
Audiobooks and podcast voiceover, accessibility features like screen reading, voice assistants and support bots, and batch voiceover for games and video are all typical TTS applications.
Related and confusable terms
Automatic speech recognition (ASR) is the inverse process: TTS generates speech from text, while ASR extracts text from speech.