Relevant Syllabi at the Intersection of Linguistics & AI

This class will introduce students to spoken language processing: basic concepts, analysis approaches, and applications. Applications include Text-to-Speech Synthesis, dialogue systems, and analysis of entrainment, empathy, personality, emotion, humor and sarcasm, deception and trust, radicalization and charisma, all using text and speech information and some visual features as well.

As technology advances, we’re relying more and more on computers and ‘virtual assistants’ to understand and interact with us using human language. In this course, we’ll talk about the methods we use to help computers process, understand, analyze, and speak our language, and we’ll talk about some of the linguistic realities that make human language so hard for computers to work with. No programming or linguistic background is needed.

We will cover ethical and social issues in NLP, especially focusing on large language models/foundation models. We’ll cover research ethics, IRB, data and participatory research, harms and their mitigation, what models we should build and who should decide, and how to apply NLP to social questions about the justice system, framing and dehumanization, disinformation and toxicity, and for counter speech.

The voice is often referred to as the first instrument, and arguably, it is the instrument with which we are all most familiar. From the very moment we are born, we are all quite adept at using our voices to
get what we want through a variety of vocalizations that become increasingly sophisticated and complex over time, ultimately shaping who we are and how we engage with the world around us. However,
despite this extreme familiarity and technical facility, the voice is also the instrument that most defies attempts at classification and theory – and especially, technological reproduction and mediation.
Artificial intelligence is drastically changing the landscape of voice technology, and with it, our relationship to ourselves and those around us. In addition to its seemingly limitless expressive
capabilities, the voice is one of the most important sites for the construction and communication of identity, both individually and societally. Together, we will explore questions of embodiment, agency,
and power through the context of the voice and its intersections with many aspects of our varied identities, including (but not limited to) race, ethnicity, gender, class, religion, age, sexual orientation,
and (dis)ability, with an eye toward deepening our understanding of how our voices influence our own and other people’s understandings of themselves in an increasingly diverse and technologicallymediated global society. This course is highly interdisciplinary in nature; readings, viewings, and listenings are drawn from a variety of sources including scientific journals, magazines and newspapers, works of art, and social media platforms like TikTok, in fields ranging from neuroscience, linguistics, and computer science to the performing arts, popular culture, and political science. Topics covered will include sonic cognition, linguistics, voice recording, transmission, and synthesis technologies, artificial intelligence and machine learning, disability, the media, and democracy and representation. Through hands-on creative and technical projects (such as creating AI-generated art, training a neural network using one’s own voice, and producing a podcast episode), writing both an analytical paper and short self-reflection statements, and in-class discussions, we will explore what it means to have, to use, to generate, and to manipulate a voice in modern society – and what the future may hold for our voices.

The everyday task of conversing, which usually seems close to effortless, requires us to carry out some very complex mental computations. Linguists, psychologists, and engineers have been trying to model this complexity in order to understand how speech and language work, leading to many innovations in technology. Speech recognition, dictation systems, grammar checkers, online searching, and dialog systems are becoming fundamental technologies. We will investigate significant aspects of human speech, and learn how scientific knowledge of this system is modeled computationally. This course will allow the student to learn about one of our species’ greatest achievements, human language, and to appreciate how that knowledge, combined with a great deal of computational creativity propels human technology. In gaining this knowledge, students will learn about how scientific innovation gives rise to a complex technological system.

An enduring aspect of the quest to build intelligent machines is the challenge of human language. This course introduces students with a background in computer science and a research interest in artificial intelligence fields to the structure of natural language, from sound to society. It covers phonetics (the physical aspects of speech), phonology (the sound-structure of language), morphology (the structure of words), morphosyntax (the use of word and phrase structure to encode meaning), syntactic formalisms (using finite sets of production rules to characterize infinite configurations of structure), discourse analysis and pragmatics (language in discourse and communicative context), and sociolinguistics (language in social context and social meaning). Evaluation is based on seven homework assignments, a midterm examination, and a final examination.

This undergraduate class looks at everyday tasks that involve natural language processing: document classification, spelling and grammar correction, dialogue systems, machine translation, cryptography and forensic linguistics. Students will get insight into how these systems work (and why it is still so difficult to do natural language processing well). We also consider social and ethical considerations such as privacy, job creation and loss due to language technologies, and the nature of machine intelligence.

Language technology has a profound influence on the way ordinary people use language. This morning, because you speak English, you may have already used voice recognition or predictive text. This course explains what language technology is and how it is available for 1% of the world’s languages. Along the way, we will attempt to answer a big pressing question for our society: Can artificial intelligence ever be inclusive of all 7000+ human languages? We will identify, describe, and explain the cross-disciplinary dimensions of this question which lie partly in social science principles of human communication (e.g. conversational turn-taking), partly in linguistics theory (e.g. how languages build words), and partly in the history of computer science (e.g. ASCII vs. Unicode). Topics include spellcheckers, translation, chatbots, and language learning aids. Topics are explored in the context of globalization, language endangerment, and the recent rapid rise of artificial intelligence.

This course introduces students to the fields of experimental phonetics, the scientific study of the sounds used in human speech, and speech processing, the study of the speech signal used in automatic speech recognition, spoken emotion detection, and other technologies. Students will learn about the physiology of speech production and perception, and they will acquire the skills necessary to accurately describe speech concepts and to analyze speech using relevant methods and tools. Turning to speech processing technology, students will explore automatic speech recognition, speech synthesis, speaker identification, and emotion recognition, and learn how our understanding of human speech production and perception informs these technologies. The course will have relevance to other disciplines in the humanities, sciences, and technical fields. This course provides theoretical foundation as well as hands-on laboratory practice.

The goal of this course is to better understand the ethical considerations that arise in the deployment of NLP technology, including how to identify people likely to be impacted by the use of the technology (direct and indirect stakeholders), what kinds of risks the technology poses, and how to design systems in ways that better support stakeholder values. Through discussions of readings in the growing research literature on fairness, accountability, transparency and ethics (FATE) in NLP and allied fields, and value sensitive design, we will seek to answer the following questions:
What can go wrong, when we use NLP systems, in terms of specific harms to people?
How can fix/prevent/mitigate those harms?
What are our responsibilities as NLP researchers and developers in this regard?

Speech (naturally spoken) is the main mode of communication between humans. Speech technology aims at providing the means for speech-controlled man-machine interaction. The goal of this course is to provide the basic concepts and theories of speech production, speech perception, and speech signal processing. The course is organized in a manner that builds a strong foundation of basics, followed by a range of signal processing methods for representing and processing the speech signal.

Speech technology includes computer speech recognition and computer speech synthesis. This hands-on
course begins with a review of basic phonetics and speech signal processing and then goes through the basic
logic and methodologies for speech synthesis and recognition. Programming expertise is not required, but
most assignments will have programming options for those interested.

Introduction to spoken language technology with an emphasis on dialog and conversational systems. Deep learning and other methods for automatic speech recognition, speech synthesis, affect detection, dialogue management, and applications to digital assistants and spoken language understanding systems.