Click Here to Register

Beginning in 1971, a team of researchers at IBM began to reorient the field of automatic speech recognition away from the study of human speech perception and language understanding, refashioning it as a problem of large-scale pattern recognition under a startling new mandate: “There’s no data like more data.” In the ensuing decades, this stunning and profound new data orthodoxy abetted a “statistical” transformation within the fields of speech recognition and later natural language processing, one that was pivotal in the rise of data-intensive algorithmic modeling as a privileged and pervasive form of computational knowledge today. This talk traces the history of automatic speech recognition and how the pursuit of language as a computational problem shaped the epistemic priorities, commercial rationale, and technical implementation of data-driven machine learning and related algorithmic practices. It tells the story of how making language into data helped make data into an imperative, opening the door for the expansion of algorithmic culture into everyday life.

Xiaochang Li is an assistant professor in the Department of Communication at Stanford University. Her research centers around the relationship between information technology and knowledge production and its role in the organization of social life. Her current book project explores the history of speech and natural language processing and how the problem of mapping communication to computation shaped the foundations of algorithmic culture. Prior to joining Stanford, She was a postdoctoral fellow in the Epistemes of Modern Acoustics Research Group at the Max Planck Institute for the History of Science in Berlin and received her PhD from the Department of Media, Culture, and Communication at NYU. Her article, “Vocal Features: From Voice Identification to Speech Recognition by Machine,” co-authored with Mara Mills, was recently awarded the 2020 Bernard S. Finn IEEE History Prize.