An Adaptive Dynamical Model of Linguistic Rhythm Sean

An Adaptive, Dynamical Model of Linguistic Rhythm Sean Mc. Lennan Proposal Defense 040406 Proposal Defense - Sean Mc. Lennan - 040406

Underlying Intuitions Somewhere between the signal and low level speech recognition, linguistic time is imposed upon real time. • Linguistic time is more relevant to speech recognition than real time. • Not all segments are created equal - certain points / intervals in the speech stream are more important for recognition than others. • Proposal Defense - Sean Mc. Lennan - 040406

What “Rhythm” Is and Is Not Rhythm - historically based primarily on the perception that different languages are temporally organized differently Three recognized rhythmic types: stress-timed (English), syllable-timed (French), and moratimed (Japanese) Rhythm implies underlying isochrony which turns out to be absent (ex. Dauer, 1983) Proposal Defense - Sean Mc. Lennan - 040406

Recent Views of Rhythm Ramus and colleagues: • examined three factors: %V ΔV ΔC %V = proportion of vocalic intervals in the signal • ΔV = variation of length of vocalic intervals • ΔC = variation of length of consonantal intervals • Proposal Defense - Sean Mc. Lennan - 040406

Recent Views of Rhythm Proposal Defense - Sean Mc. Lennan - 040406

Recent Views of Rhythm Proposal Defense - Sean Mc. Lennan - 040406

Recent Views of Rhythm Proposal Defense - Sean Mc. Lennan - 040406

Rhythm and Segmentation Cutler and Colleagues study the question of how rhythm type impacts on the segmentation of words from the speech stream • implication being that a naïve listener (i. e. an infant) uses rhythm as a bootstrap for early stages of acquisition • Showed that boundaries are imposed on the speech stream in a rhythm-class-appropriate manner • Proposal Defense - Sean Mc. Lennan - 040406

The Proposed Model • hopefully a bridge between Cutler et al and Ramus et al - why should %V ΔV ΔC impact on segmentation? • can a naïve adaptive model responsive to %V ΔV and ΔC produce behavior consistent with segmentation based on rhythm-type? Proposal Defense - Sean Mc. Lennan - 040406

The Proposed Model - Where Finding salient points in the signal: Proposal Defense - Sean Mc. Lennan - 040406

The Proposed Model - How Much • %V ΔV and ΔC need two points to be consistently tracked: vocalic onsets and offsets Proposal Defense - Sean Mc. Lennan - 040406

The Proposed Model - How Much Use these spikes to drive an adaptive oscillator • Unlikely to entrain but will make predictions • Proposal Defense - Sean Mc. Lennan - 040406

The Proposed Model - How Much The accuracy of prediction will be a measure of ΔC and ΔV • Difference in the period will be a measure of %V • Proposal Defense - Sean Mc. Lennan - 040406

The Proposed Model - How Much Voc ΔC ΔV Cons Proposal Defense - Sean Mc. Lennan - 040406

Proposed Model - How Much Proof of Concept - Periodic Signal Proposal Defense - Sean Mc. Lennan - 040406

Proposed Model - How Much Proof of Concept - Aperiodic Signal Proposal Defense - Sean Mc. Lennan - 040406

The Proposed Model - How Much • Attentional window size (hopefully) would correlate with rhythm type and would predict different potential segmentation boundaries Proposal Defense - Sean Mc. Lennan - 040406

The Proposed Model Predictions, questions, and other benefits: • • • consistent with the correlation between rhythmic type and consonant cluster complexity consistent with ambisyllabicity perhaps attractor states predict categorical differences suggests manner in which to manipulate tasks to force effects particularly with respect to speaking rate and rhythmic priming single language-independent mechanism Proposal Defense - Sean Mc. Lennan - 040406