The latest honor for Shri Narayanan speaks to his “ingenuity, as well as his leadership in solving multidisciplinary challenges,” says USC President Carol Folt. (Photo/Angel Ahabue)
Shrikanth “Shri” Narayanan’s latest major award is personally a big one for him. It’s the highest technical honor in his engineering field. What makes it even more special is that he knew the man who inspired the Institute of Electrical and Electronics Engineers (IEEE)’s James L. Flanagan Speech and Audio Processing Award.
Narayanan and Flanagan crossed paths at the legendary AT&T Bell Laboratories in Murray Hill, New Jersey. It had been founded in 1925 as the Bell Telephone Laboratories, run jointly by Western Electric and AT&T. Flanagan joined Bell in 1956 and conducted groundbreaking research for decades before retiring in 1990. Narayanan started there as an intern in the summer in 1992, having just finished his master’s in electrical engineering at UCLA.
“It was such an open environment of intellectual energy and collegiality,” Narayanan said. “Here I am, 30 years since then, and I’m still following the ideas I was exposed to there.”
Narayanan has been at USC since 2000 and is now a University Professor; the Niki and Max Nikias Chair in Engineering; and Professor of Electrical and Computer Engineering, Computer Science, Linguistics, Psychology, Pediatrics, and Otolaryngology-Head & Neck Surgery. He holds academic appointments in multiple schools at the university, and his work spans a multitude of disciplines, from human speech science and language technology to behavioral machine intelligence.
“This honor speaks to Shri’s ingenuity, as well as his leadership in solving multidisciplinary challenges,” said President Carol Folt. “He’s amazing: His ability to collaborate with others and break down silos drives his breakthroughs in speech processing and many other areas, and is a strength he brings to everything he does. His personal connection to the award’s namesake makes it even more remarkable, and I congratulate him on this well-deserved recognition.”
“Shri epitomizes the convergence of disciplines like no one else in his field,” USC Viterbi School Dean Yannis C. Yortsos said. “He creates new building blocks in the technology-humanity double helix for the collective benefit. We are fortunate to have him as our colleague and celebrate with joy all his achievements.”
Narayanan speaking at the Interspeech conference in Dublin in August 2023. (ISCA/Interspeech)
The underlying theme of Narayanan’s work centers on how humans communicate, and what we can learn from those methods to improve the human condition and experience. As a twentysomething intern at Bell Labs, however, he had no inkling of the path he would take. “I was working on video processing in my first internship, to transmit it efficiently over the network, a project where the big idea was to be able to see the person when you’re talking to them. In those days, when you spoke on the phone, it was just a handset where you heard them. You called them up, but you didn’t get to see them. There was no FaceTime or Zoom.”
A renowned researcher in the department Flanagan had led, Man Mohan Sondhi, asked Narayanan whether he might consider pursuing doctoral research in speech and audio processing. “He said there’s beautiful science and engineering: from unpacking how humans use their vocal instrument to produce these sounds that turn thought into signals that people process, and in turn respond back. He said there’s an amazing wealth of research questions in understanding speech and in creating technologies for people. He also said UCLA, where I was in graduate school, has an amazing linguistics department with leading researchers in speech science. So I truly got my inspiration and start in speech and audio processing right there at Bell Labs.”
Narayanan earned his Ph.D. at UCLA, returned to AT&T Bell Labs in 1995, then went into academia full-time. But the friendships he forged in the early 1990s in Murray Hill have never stopped echoing. After retiring from Bell Labs, Flanagan joined the faculty at Rutgers University, and he counseled a visiting Narayanan about his thesis work, as well as his early research. Flanagan died in 2015. (Bell Labs is now Nokia Bell Labs.) Sondhi, who passed away in 2018, served on Narayanan’s Ph.D. committee. Flanagan’s adviser, Ken Stevens, was Narayanan’s Ph.D. adviser, UCLA’s Abeer Alwan’s adviser, and a mentor for Narayanan as well. (Stevens was a co-winner of the first Flanagan Award, with Gunnar Fant, in 2004. See a list of all the past winners.)
“To be honored with such a recognition, where all your heroes are, is very humbling,” Narayanan said.
The award recognizes his “contributions to speech communications science and technologies for inclusive human-centered engineering” and will be presented in April 2025 at the IEEE International Conference on Acoustics, Speech, and Signal Processing in Hyderabad, India. Narayanan was born to the north, in New Delhi, and earned his bachelor’s in electrical engineering from the College of Engineering-Guindy in Chennai.
Narayanan’s emphasis on collaboration across different scientific fields, and across continents, has brought him and his research group (the Signal Analysis and Interpretation Laboratory, or SAIL) widespread acclaim. He received a Guggenheim Fellowship in 2022, and earlier this year he was named the first USC vice president for presidential initiatives by Folt, supporting USC’s “moonshot” priorities, including the realms of computing, health and sustainability.
His work continues during a time of unprecedented leaps forward in human communication: The “telephony”-inspired speech and audio processing he worked on at Bell Labs have given way to tools that use artificial intelligence to translate what we say, as well as interpret what we mean and get at the heart of who we are.
“It’s not just translating words now, right? It’s capturing who you are and it’s communicating with all the intonation and all the affect,” he said. “How do you capture and not lose that in translation? What is unique about speech and the spoken language we use is it provides means for expressing ourselves, not just our intent but our emotions and our desires. And it’s also very interactive: It captures not just thought but is shaped by who you are, your identity, your culture. It captures and offers a valuable window into your health status.”
“To create machines to have that ability to understand human speech and language is a profoundly fascinating area of interdisciplinary inquiry. At USC over the last 25 years, I’ve had the good fortune of studying all these things, and the journey has been joyful.”
Published on September 13th, 2024
Last updated on September 16th, 2024
