Book

Automatic Speech Recognition: A Brief History of the Technology Development

B. H. Juang and Lawrence R. Rabiner, Elsevier Encyclopedia of Language and Linguistics (author copy, UC Santa Barbara), 2005

Read the original

Read on October 16, 2026

What it is: A history of automatic speech recognition by two of its leading researchers, from the 1952 Bell Laboratories digit recognizer through statistical methods to commercial dictation systems.
How much weight it carries: Written by pioneers of the field and widely used as a reference. It ends in the mid-2000s, before neural network recognizers.

Facts taken from this source

In 1952, Davis, Biddulph and Balashek of Bell Laboratories built a system that recognized isolated spoken digits from a single speaker.
“In 1952, Davis, Biddulph, and Balashek of Bell ... built a system for isolated digit recognition for a single speaker” (Section on early systems)
Juang and Rabiner write that IBM and AT&T Bell Laboratories were two different schools of thought on commercial speech recognition in the 1970s, and that IBM's effort was led by Fred Jelinek.
“representing two different schools of thought as to the applicability of automatic speech recognition systems for commercial applications. IBM’s effort, led by Fred Jelinek,” (Section on 1970s research directions)
Juang and Rabiner write in 2005 that most practical speech recognition systems were based on statistical methods developed in the 1980s.
“Today, most practical speech recognition systems are based on the statistical framework and results developed in the 1980’s, with significant additional improvements in the 1990’s.” (Section 5, technology directions in the 1980s and 1990s)
Juang and Rabiner write that Hearsay-II and BBN's HWIM did not meet the DARPA program's performance goal when it ended in 1976.
“Neither Hearsay-II nor HWIM (Hear What I Mean) met the DARPA program’s performance goal at its conclusion in 1976.” (Section on the DARPA Speech Understanding Research program)
IBM's Tangora, a research system meant to act as a voice-activated typewriter, had to be trained by each individual user, so it was speaker-dependent.
“the typewriter had to be trained by each individual user” (Section on the IBM and Bell Laboratories research programs)

Pages that rest on this source

Book

Automatic Speech Recognition: A Brief History of the Technology Development

B. H. Juang and Lawrence R. Rabiner, Elsevier Encyclopedia of Language and Linguistics (author copy, UC Santa Barbara), 2005

Read the original

Read on October 16, 2026

What it is: A history of automatic speech recognition by two of its leading researchers, from the 1952 Bell Laboratories digit recognizer through statistical methods to commercial dictation systems.
How much weight it carries: Written by pioneers of the field and widely used as a reference. It ends in the mid-2000s, before neural network recognizers.

Facts taken from this source

In 1952, Davis, Biddulph and Balashek of Bell Laboratories built a system that recognized isolated spoken digits from a single speaker.
“In 1952, Davis, Biddulph, and Balashek of Bell ... built a system for isolated digit recognition for a single speaker” (Section on early systems)
Juang and Rabiner write that IBM and AT&T Bell Laboratories were two different schools of thought on commercial speech recognition in the 1970s, and that IBM's effort was led by Fred Jelinek.
“representing two different schools of thought as to the applicability of automatic speech recognition systems for commercial applications. IBM’s effort, led by Fred Jelinek,” (Section on 1970s research directions)
Juang and Rabiner write in 2005 that most practical speech recognition systems were based on statistical methods developed in the 1980s.
“Today, most practical speech recognition systems are based on the statistical framework and results developed in the 1980’s, with significant additional improvements in the 1990’s.” (Section 5, technology directions in the 1980s and 1990s)
Juang and Rabiner write that Hearsay-II and BBN's HWIM did not meet the DARPA program's performance goal when it ended in 1976.
“Neither Hearsay-II nor HWIM (Hear What I Mean) met the DARPA program’s performance goal at its conclusion in 1976.” (Section on the DARPA Speech Understanding Research program)
IBM's Tangora, a research system meant to act as a voice-activated typewriter, had to be trained by each individual user, so it was speaker-dependent.
“the typewriter had to be trained by each individual user” (Section on the IBM and Bell Laboratories research programs)