A component containing data and, optionally, executable code that helps to shape the sound of synthesized speech.