The central idea of this paper is the existence of a nonsemantic component in the information conveyed by speech. This component is composed of bits of nonsemantic information that are part of the structures formed by their quantitative and elective (carrying directed substitution) relations. Such structures serve as specific parameters of word meanings used in speech and are utilized as language description units. The nonsemantic component of such word classes as verb, noun, interjection and numeral is a subject of interest of the paper as well as the role of this component in sentence formation. The importance of nonsemantic information is that it acts as a background, a foundation of sentence grammar. The paper deals with such issues as the sign and its relation to meaning in speech, meaning and information.