[Skip to Navigation]
August 1960

Combinations of Distortion in Speech: The Twenty-Five Per Cent Safety Factor by Multiple-Cueing

Author Affiliations

New London, Conn.
U.S. Naval Medical Research Laboratory.

Arch Otolaryngol. 1960;72(2):227-232. doi:10.1001/archotol.1960.00740010232014

Introduction  Many writers have noted the very wide latitude possible to a speaker in varying his fundamental frequency and timbre, rate of talking, accent, nasality, etc., before his speech is degraded in intelligibility. In the laboratory one can remove all frequency cues by clipping, discard or distort major frequency regions, interrupt periodically, mask very nearly completely with noise, and do violence to the signal in many other ways; nevertheless, intelligibility will not be lost and often remains amazingly high.The generic term redundancy is now in common use to denote the idea that if a word is missed in a sentence there are enough cues and restrictions furnished by syntax and context to give the listener a very good idea of the missing item. The word redundancy has been extended by some writers to the case of multiple cueing, as when either of two frequency regions will suffice to carry