DYMOS: Dynamical models of speech

A conference exploring new frontiers in dynamical models of speech. The event loosely follows on from our 2024 satellite workshop on Variance and invariance at LabPhon 19.

Location and date

The event is co-located at the 2025 LSA Linguistics Institute and coincides with the class Dynamics of speech production, taught by Jason A. Shaw and Michael C. Stern.

The conference will take place in Straub Hall at University of Oregon:

Registration and further information

Please note that registration has now closed.

There is helpful information about travel, accommodation, etc. at this link (thanks to the organizers of Usage-Based Approaches to Phonology for compiling this - please check out their conference too!).

The conference will be hosted using the same facilities as the LSA Institute. You can find more information at the LSA Institute website.

Please contact Sam Kirkham if you have any questions.

Program

This is currently a draft program and all timings are indicative and subject to change.

Saturday

Time Speaker Title
0900 Arrivals -
0915 Sam Kirkham Introductions
0930 Adamantios Gafos (virtual) Kinematic and topological features of speech and speech-like action
1030 Coffee break -
1100 Scott Nelson Using model theory and function types to formally relate symbolic and dynamic theories of speech
1130 Alice Turk From phonological symbols to articulation: A 3-component model of speech production
1230 Lunch -
1400 Rachel Walker & Michael Proctor Dynamics of vocalic articulations in syllable rhymes: Sesquisyllables in Australian English
1430 Michael Stern Prosodic influences on the dynamics of articulatory movement and control
1530 Coffee break -
1600 Manasvi Chaturvedi Modeling the dynamics of online vowel correction
1630 Jason Shaw Some limitations on deriving phonological patterns from neural field dynamics
1730 End of program -

Sunday

Time Speaker Title
0900 Melissa Redford The dynamic coordination of breathing and language in speech
1000 Coffee break -
1030 Carina Ahrens Social influences on articulatory targets
1100 Sam Tilsen On the choice of autonomy in dynamical models of speech
1200 Lunch -
1330 Khalil Iskarous Articulatory phonology in the age of neural syntax and reinforcement learning
1430 Sam Kirkham Discovering dynamical laws from data
1530 Coffee break  
1600 Open discussion and future plans -
1730 End of program -

Workshop speakers

Organisers:

Description

A fundamental topic of linguistic inquiry concerns the nature of phonological cognition. Dynamical models of speech – most prominently Articulatory Phonology (Browman & Goldstein 1992) / Task Dynamics (Kelso, Saltzman & Tuller 1986) – have proven great successes in modelling the dynamics of coarticulation (Fowler 1980), prosody (Byrd 2003; Iskarous et al. 2024), long-range processes (Gafos & Benus 2006; Tilsen 2019), and neural dynamics of speech planning (Roon & Gafos 2006; Tilsen 2022; Stern & Shaw 2023). At the same time, there remain a range of unresolved issues and critiques, including the nature of timing control, the relations between symbolics and dynamics, the emergence of phonological structure, and what counts as a necessary condition of a dynamical theory. The conference aims to assess the state-of-the-art and future research directions in dynamical models of speech, with a focus on the following topics:

  1. Emergence and change in dynamical phonological categories
  2. Dynamics, variation, and linguistic universals
  3. The status of spatial and temporal variation in task dynamic models
  4. Autonomous and non-autonomous gestural models
  5. Neural dynamical models of speech planning
  6. Dynamical models of production-perception
  7. The relationship between task dynamic and biomechanical models
  8. Feedback controls
  9. Systems-level dynamics
  10. Philosophical foundations

This two-day conference will bring together researchers in dynamical models of speech, featuring plenary talks from invited speakers, contributed talks, and space for discussion.