The generation of an sound waveform of human speech from a textual or phonetic description. See also speech recognition.