Task 9 speech perception Flashcards
Phoneme
: the shortest segment of speech that, if changed would change the meaning of the word
Speech segmentation
the perception of individual words in a conversation
o Knowledge of meaning of words
o Some sounds are more likely to appear together within a word and other sounds are more likely to be separated by the space between two words
o Transitional properties: the chances that one sound will follow another sound
Acoustic signal (production)
the movement of structures within the vocal apparatus, which produce patterns of pressure changes in the air
Articulators (production)
due to moving of the articulators the vocal tract is shaped, includes tongue teeth, jaw and soft palate
Vocal tract (production)
articulation, nasal and oral tract combined
Vocal folds (production)
phonation, producing sound (vocal cords)
Vowels
are produced by vibration of the vocal cords
Formats (procduction)
the frequencies at which these peaks (pressure change) occur
Consonants
produced by closing or constriction of the vocal tract
Format transition (production)
rapid shifts in frequency preceding or following formats, consonants
Motor theory of speech perception
1. Hearing a particular speech sound activates motor mechanisms controlling the movement of the articulators that are responsible for producing sound
2. Activation of these mechanisms enables us to perceive the sound
Recognition
o Easier to perceive phenoms which are aligned in a meaningful context around 8% faster recognition
Phonemic restoration effect
when a word is presented with a cough at a certain point and the letter was taken out the subjects could not tell at which point the cough was and didn’t recognized that the letter was actually missing, can be influenced by certain situations (Top down processing)
Works best when the masking sound has the same frequency as the original letter
Categorical speech perception
occurs when stimuli that exist along a continuum are perceived as divided into discrete categories
o Voice onset time: the time delay between when a sound begins and when the vocal cords begin vibrating
o Phonetic boundary: When the perception from e.g. da to ta changes (at 35ms) we perceive the phenoms different on each site of the boundary
Multimodal perception
speech perception I multimodal that states that our perception can be influenced by visual information from a number of different senses
McGurk effect
when we see a video where someone moves the lips according to ga-ga and we hear ba-ba we will perceive da-da
Audiovisual speech perception
the influence of vision on auditory perception
o Might be neural mechanism behind because lip reading activates a different part of the audio-cortex also activates FFA when we know the voice of the person
Voice area (cortical)
Are in the superior temporal sulcus, that reacts more to voices than to other sounds (part of ventral pathway)
Voice cells (cortical)
detected in monkeys temporal lobe, they respond more strongly to monkey calls than to calls of other animals or to “non-voice sounds”
What pathway
ventral pathway starting at the interior auditory cortex → interior temporal lobe → frontal cortex that responsible for recognizing speech (Brocas)
Where/How pathway
dorsal pathway starting at the posterior auditory cortex → goes to parietal lobe → motor cortex that that is responsible for linking the acoustic signal to the movements used to produce speech
Broca’s aphasia
They have laboured and stilted speech and can only speak in short sentences, they are still capable of understanding what others say (frontal lobe above lateral fissure)
Wernicke aphasia
They can speak fluently but what they say is extremely disorganized and not meaningful. These patients have difficulties with understanding what others say (temporal lobe)
o Word deafness: an extreme form of Wernicke apahsia in which they cannot recognize words, even though the ability to hear pure tones remains intact
Damage to parietal lobe
problems with discriminating syllables, some can still understand words
Indexical characteristics
carries information about the speaker such as age, gender, place of origin, emotional state and if they are being sarcastic or serious
o Two levels of information about a word: 1: its meaning 2: characteristics of the speakers voice
Audiovisual mirror neurons
These neurons fi re when a monkey carries out an action that produces a sound (like breaking a peanut) and when the monkey hears the sound (the sound of a breaking peanut) that results from the action
Spectogram (speech)
time is length frequency is height, and amplitude is colour. Vowels have greater amplitude
Coarticulation
the sounds are made differently depending on which sound precedes another (de and du) d is basically the same but the format is different, perceptual constancy
Spectral contrast
Same sound different intensities, linked to brightness and contrast in vision, some contrasts are more likely to occur together, makes it easier to discriminate phonemes/words