Ben’s success in translating Morse code led to an early effort in automatic machine recognition of speech using an early advanced digital computer (TX2) at Lincoln Laboratory. The focus of this early work was on the precision detection of speech pitch, an important component of early speech-compression devices, called Vocoders, as well as early speech-recognition systems.
In 1965, Ben was invited to work with MIT Professor Ken Stevens on the theory of digital-signal processing, which became a central capability in many subsequent applications, including speech recognition, radar, sonar, communications, seismology, and biologic systems. While on the MIT campus, Ben taught what is believed to be the first course on this subject. His pioneering research continues to have an enormous impact on modern electronic devices and systems. In 1969, he published (with Charles Rader) Digital Processing of Signals, the first treatise, and seminal text, on this topic. He then focused his attention on human-speech perception and related topics. In 1975, he published (with Lawrence Rabiner) a comprehensive textbook on digital-signal processing, Theory and Applications of Digital Signal Processing. Finally, in 2000, he published (with Nelson Morgan) Speech and Audio Signal Processing.
In his later career, Ben played a leading role at Lincoln Laboratory in the development of customized, very-high-speed digital-signal processors designed for specialized fast-Fourier transform operations, and digital filtering. These designs led to a number of important applications, including coherent Doppler radar processors, very-high-speed packet speech processors, and clutter rejection in air traffic control radar. These specialized computational processors achieved speeds far in excess of the speed of conventional, nonspecialized processors. Ben published more than 20 significant papers on applications of digital-signal processors.
In 1988, at the age of 65, he retired from research at MIT Lincoln Laboratory, although he continued to conduct research on the MIT campus on speech synthesis, speech recognition, human auditory systems, and related areas. He also taught courses on these topics at the University of California, Berkeley, for several years.