JLP374 - wk 1-4 - Speech Perception 3 issues: Segmentation:...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Speech Perception 3 issues: Segmentation: Unlike exaggerated speech, natural speech has no real pauses, hard to line up start/stop in spectogram Lack of invariance problem: Individual sounds have blurry boundaries and coarticulation issues (/s/ in seep different from /s/ in soup in spectrogram due to ‘anticipation’ of the remainder of the word) Speaker Variability problem: Male vs female, child vs adult – all different. Like searching for a ‘dog’ in a group of Jack Russells and Dobermans. McGurk Effect: ‘baba’ sound heard as ‘dada’ when ‘baba’ is said with ‘dada’ mouth movements! Compare and Contrast: Cohort Theory: Features: Initial sounds define ‘cohorts’ (possible words) and each successive phonological info narrows Each word has ‘resting activation level’ – as word unfolds, compatible nodes activated and incompatible are deactivated Problem: does not work if the initial sound is ‘barred’ by noise! Trace Model: Features: Continuous activation (each successive sound unit in unfolding word provides independent source of evidence about word identity , regardless of what sounds came before) Uses top-down information (knowledge/expectations) along with bottom-up (sensory) information
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/17/2008 for the course JLP 374 taught by Professor Pittman during the Fall '08 term at University of Toronto- Toronto.

Page1 / 3

JLP374 - wk 1-4 - Speech Perception 3 issues: Segmentation:...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online