Oral History of Raj Reddy

Описание: Interviewed by Hansen Hsu on 2021-04-14 in Pittsburgh, PA X9458.2021
© Computer History Museum

Raj Reddy was born on June 13, 1937, in the Telugu-speaking farming village of Katur, in the state of Andra Pradesh, India. Reddy was sent to a boarding school from the age of 9, later attending Loyola College and then the College of Engineering, Guindy in the city of Chennai (formerly Madras). In 1958 he began graduate studies in civil engineering at the University of New South Wales in Australia. His advisors there, Stan Hall and Bob Woodhead, introduced him to computers in the form of an English Electric DEUCE Mark II. After finishing his Master’s degree Reddy got a job working for IBM Australia doing systems programming. It was during these years that Reddy read about the emerging field of artificial intelligence, and decided to get a Ph.D. Reddy was accepted at Stanford in 1963 and worked under AI pioneer John McCarthy. Because Stanford AI Lab had a new PDP-1 minicomputer with an analog-to-digital converter, and due to his own multi-lingual background, Reddy began the project that would come to define his career: continuous speech recognition. Reddy was, along with Bill McKeeman, the first to receive a Ph.D. from Stanford’s new Computer Science department in 1966. He stayed on at Stanford as an assistant professor from 1966-1969, continuing to work on speech recognition, demonstrating its use to give commands to a robotic arm in 1968.

In 1969 Reddy moved to Carnegie Mellon University where he would spend the rest of his career. 1971 marked the beginning of DARPA’s Speech Understanding Research (SUR) program, which funded several institutions’ speech programs including those supervised by Reddy at CMU. Because Reddy was willing to let his students pursue multiple ideas under his guidance, CMU produced not one but three speech recognition systems during this time: HEARSAY I/II, DRAGON, and HARPY. DRAGON, by James and Janet Baker, was the first system to use the statistical “Hidden Markov Model” (HMM) technique, and led to the technology’s commercialization at their company, Dragon Systems, in the 1980s. HMMs became the dominant technology for speech recognition. Later that decade, a new round of DARPA funding would lead Reddy’s student Kai-Fu Lee to develop SPHINX, the first speaker-independent system.

In the 1970s, Reddy also became a consultant at Xerox PARC, leading to CMU developing an early laser printer prototype. In 1979, Reddy cofounded and became the first Director of the Robotics Institute at CMU, the first academic department in the US dedicated to robotics. In the 1980s, Reddy also became involved in Le Centre Mondial, a French government project to spread computing in the developing world, for which Reddy received the Legion of Honor. Reddy became Dean of the School of Computer Science (SCS) at CMU from 1991-1999. As Dean, Reddy oversaw the creation of several new departments: Human Computer Interaction Institute, Language Technology Institute, and the Department of Machine Learning. Reddy also served as a catalyst for CMU’s expansion to its Doha, Qatar campus, offering both Computer Science and Business degrees.

In the last couple of decades, Reddy has shifted his work towards expanding access to information technology to the benefit of more people, including those in the developing world, in keeping with his belief that education and literacy belong to all. This effort includes the founding of Rajiv Gandhi University for Knowledge Technologies (RGUKT) in his home state of Andra Pradesh in India, whose mission is to bring education to the brightest of students in each rural village in the state.

Note: Transcripts represent what was said in the interview. However, to enhance meaning or add clarification, interviewees have the opportunity to modify this text afterward. This may result in discrepancies between the transcript and the video. Please refer to the transcript for further information - http://www.computerhistory.org/collec...

Visit computerhistory.org/collections/oralhistories/ for more information about the Computer History Museum's Oral History Collection.

Catalog Number: 102792183
Lot Number: X9458.2021