Speech synthesis

Speech synthesis is the artificial production of spoken language from text or other input. It ranges from rule-based systems to models that learn speech patterns from recorded voices.

Text-to-speech: Technology that converts written text into spoken audio. It is the most common form of speech synthesis.
Phoneme: A language-specific sound category that distinguishes words. Many synthesis systems represent pronunciation as sequences of phonemes.
Wolfgang von Kempelen: An inventor whose eighteenth-century speaking machine mechanically imitated parts of human speech production. His machine is an early documented attempt to synthesize speech.
Screen reader: Assistive software that presents digital content through speech or other output. Speech synthesis lets screen readers voice text and interface information.
Augmentative and alternative communication: Methods and tools that supplement or replace speech for people with communication disabilities. Speech synthesis is one way these systems turn chosen messages into spoken language.
Text normalization: The conversion of written forms such as numerals and abbreviations into spoken-word sequences. It determines how symbols in an input text are read aloud.
Formant: A concentration of acoustic energy at a frequency shaped by vocal-tract resonance. Formants are central to modeling vowel quality in speech.
Kempelen speaking machine: An eighteenth-century apparatus that produced speech-like sounds through manipulated airflow and resonators. It made speech synthesis a physical mechanism before electronic computation.
Audiobook: A recording of a book or other text intended for listening. Synthetic narration can produce spoken editions without a human recording session.
Speech privacy: The protection of spoken information from unwanted access or disclosure. Generated voices can expose private content when systems speak in shared spaces.

Linked from 14 pages