task 9 Flashcards
(25 cards)
acoustic signals
speech sounds are made of acoustic signals=patterns pf pressure changes in the air produced by movements of structures in the
vocal apparatus (lungs, vocal cords, tongue)
speech in steps
- Air pushes up from the lungs
- It poses though the larynx and vocal cords
- It enters the vocal tract (nasal and oral cavities, and pharynx)
- Articulators ( tongue, lips, teeth, jaw, mouth, soft palate move, which changes the shape of the vocal tract.
- This changes the airflow and resonant frequencies I the vocal tract which produces specific speech sounds.
production of vowels
vowels: produced by the opening of the vocal tract
*changes of shape of the vocal tract modifies the resonant frequency and produces formants
formants(=peaks of pressure at specific frequencies)
each vowel has specific formants: F1 has the lowest frequency, F2 has a higher one
production of consonants
produced by closing the vocal tract
->associate with formant transitions= rapid shifts in frequency preceding or following formants.
classified by:
*manner of articulation- how the articulators interact with each other when making a speech sound
*place of articulation- location of the articulation
*vocing- whether or not the vocal cords are vibrating
sound spectrogram
3D visual representation of the acoustic signals (spectrum of frequencies)
- time on x-axis, frequency on y-axis and intensity/ amplitude in color (darker areas indicate greater intensity)
phenome
smallest units of speech that, if changed, changes the meaning of a word
*dont refer to letters but to the distinctive speech sounds pf vowels and consonants (since they can have different pronunciations)
variability problem
challenge of recognizing and understanding speech because specific phenomes can be associated with multiple different acoustic signals. This is due to:
*contextual cues: coarticulation- overlap between articulation of neighboring phonemes
*since articulators are constantly moving while talk, the shape of the oval tract associated with particular phenomenon is influenced by the sounds that precede and follow that phenome.
*Perceptual constancy- we perceive the sound of particulate phenomes as the same even though the acoustic signal of the phoneme changes.
pronunciation as on of the variability problems
Between-speaker difference- difference in how multiple people pronounce words.
* due to pitch, speed, accent
Within speaker differences- difference in how same person pronounces words
* due to the context: speaking in front of siblings vs teacher
motor theory of speech perception
based on the facts that motor commands (movement of the vocal cord) are directly linked to phenoms (1:1) relationship
Hearing speech sounds triggers the motor processes in the listener that are associated with producing the sound. Our brain uses knowledge of producing the sound to understand what is being said.
*but it doesn’t explain how people with Brian damage, disabling their speech motor system, can still perceive and understand speech
*But it doesn’t explain how infants understand speech before learning how to speak
categorical perception
we perceive speech base on the categories to which we attribute the sounds that compose the words
*Voice onset time (VOT): time delay between when a sound begins and when vocal cords begin vibrating
ex: /da/ has a shorter VOT than /ta/
*Phonetic boundary:VOT transition point at which perception changes from one category to another.
stimuli on the same side of the phonetic boundary are perceived as the same category -> perpetual constancy
phonemic restauration effect
effect when missing sounds are not noticed because we know the meaning of the word of sentence, so our brain fills the missing sound. It’s due to top-down processing: prior knowledge creates expectations.
speech segmentation
perception of individual words in conversation
even though the acoustic signal for a spoken sentence is continuous, we perceive breaks between words, because we know the meaning of the words.
transitional probability
chance that one sound will follow another
*statistical learning= process of learning about traditional probabilities, especially when leaning a new language.
McGurk effect
speech perception is multimodal -it can be influenced by information from other sense, like vision, McGurk effect-seeing a person’s lips make movements for one sound influences how we perceive the actual sound that reaches our ears.
->when we see someone saying a phenomenon different to the one we hear, our brain matches the sound we perceive to the one we see.
dual stream model of speech perception
speech perception in the bairn: 2 pathways to perceive speech
ventral (what pathway)
from anterior auditory area to the frontal lobe.
*responsible for speech comprehension
dorsal (where) pathway
from the posterior auditory area to the parietal lobe and motor cortex
*responsible for liking acoustic signals to movements used to produce speech
speech production (frontal lobe)
Broca’s area
primary motor cortex
Premotor cortex
dorsal pathway
speech comprehension (temporal lobe)
wernicke’s area
primary auditory cortex
ventral pathway
interconnection
but different area work together in an interconnected network
Problems in speech
aphasia, broca’s aphasia, wernicke’s aphasia, word deafness
aphasia
language disorder due to damage of the brain area that controls language expression and comprehension
broca’s aphasia
due to damage to broca’s area in the frontal lobe-> impairs speech
difficulties in forming complete sentences but are able to underhand some types of sentences
word deafness
inability to recognize words, even though they can still hear pure tones
