24th March 2021 11:20 AM

PhonDAWG - Phonetics Lab Data Analysis Working Group

Sam will give a practice talk.   Note that the talk is 60 minutes. It's fine if people have to leave before it's finished.

25th March 2021 04:30 PM

Linguistics Colloquium Speaker: Dr. Robert Ladd of Edinburg University

The Department of Linguistics proudly presents Professor Emeritus Robert Ladd from the University of Edinburgh.  Dr. Ladd is a Cornell and Linguistic Department alumnus (MA Linguistics 1972; PhD Linguistics 1978).

Dr. Ladd with give a talk titled:  "It's prominence, but not as we know it" (Adventures in the phonology and phonetics of stress)

Abstract:

A simple notion of “prominence” has taken centre-stage in recent phonetic work on how prosodic features are used to signal pragmatic and syntactic properties.  A good deal of current experimental work attempts to identify the range of phonetic and other features that distinguish words judged “prominent” from other words. 

But assumptions about the relation between phonetic prominence and syntactic/pragmatic importance (focus, informativeness, etc.) - including the assumption that a language will exhibit such a relation - are Eurocentric and now fairly uncontroversially wrong. More importantly, prominence is ill-defined, and little attempt is made to relate the recent findings of phonetic investigation to theoretical ideas about “stress”. 

If we think about stress in terms of abstract strength relations in a hierarchical structure - i.e. in terms inspired by Liberman’s original proposal for metrical phonology and his notion of lawful “tune-text association” - we can to go beyond trying to account phonetically for impressionistic “prominence” and can recognize genuine typological differences in the way languages use suprasegmental phonetic resources.

26th March 2021 09:55 AM

Dr. Robert Hawkins will talk on "Coordinating on meaning in communication"

This week in Computational Psycholinguistics Discussions (C.Psyd), we're excited to host invited speaker Dr. Robert Hawkins of Princeton. Talk details below.

Title: Coordinating on meaning in communication

Abstract: Languages are powerful solutions to coordination problems: they provide stable, shared expectations about how the words we say correspond to the beliefs and intentions in our heads. However, in an non-stationary environment with new things to talk about and new partners to talk with, linguistic knowledge must be flexible: old words acquire new ad hoc or partner-specific meanings on the fly.

In this talk, I'll share some recent work investigating the cognitive mechanisms that support this balance between stability and flexibility in human communication, which motivates the development of more adaptive, interactive language models in NLP.

First, I'll introduce a computational framework re-casting communication as a hierarchical meta-learning problem: community-level conventions and norms provide stable priors for communication, while rapid learning within each interaction allows for partner- and context-specific common ground. I'll evaluate this model using a new corpus of natural-language communication in a communication task where participants are grouped in small communities and take turns referring to ambiguous tangram objects and describe how we scaled up this framework to neural architectures that can be deployed in real-time interactions with human partners.

Taken together, this line of work aims to build a computational foundation for a more dynamic and socially-aware view of linguistic meaning in communication.

Bio: Robert Hawkins is currently doing his post-doc at Princeton. He spent his undergraduate years at Indiana University and did his PhD at Stanford. Broadly, Robert works at the intersection of cognitive science and machine learning and is interested in the cognitive mechanisms that allow people to flexibly coordinate and collaborate with one another, particularly those that allow social conventions and norms to emerge.

26th March 2021 12:20 PM

Phonetics Lab Meeting

We will continue our discussion of Xu and Xu 2005, focusing on the statistical analysis and interpretation.