Speech perception - week 11 (Nicholas)

Question 1

Speech is variable

Answer 1

Every word takes on a different acoustic shape each time it is uttered. This is due to…
-speaker (vocal track size, regional accent, socio-economical tier)
-articulation rate (4/5 syllables/sec in sentences)
-prosody (the music of speech: rhythm, melody, amplitude)
-mode: voiced, whispered, creeky, etc…
-coarticulation: individual phonemes influenced by preceding/upcoming segments (i.e., regressive and progressive assimilation). e.g., ‘an obb pronunciation’

Question 2

Speech is quasi-continuous

Answer 2

Speech has no unique/systematic way to flag word boundaries (i.e., there is rarely silence between two words).
Note. Short silences (around 100 ms) typically correspond to the vocal tract closing to produce so-called plosive (or STOP) consonants, in ‘pocket’.

Question 3

Speech is lexically ambiguous

Answer 3

Words are made out of a limited number of sounds and syllables. Therefore, embedded words are everywhere, inside other words.
e.g., ‘captain’. (cap what?)
Note. Ambiguity also arises due to straddling words, as soon as we put two words together.
e.g. “clean ocean” (notion?)

Question 4

Speech is audiovisual

Answer 4

Visual information given by the lips and adjacent areas of the face about articulation is integral to speech perception, when available.

Question 5

McGurk and McDonald’s (1976) fusions

Answer 5

/ga/ (vision) + /ba/ (audition) = /da/ (perception)

Question 6

What kind of information does the listener need for identifying words?

Answer 6

Phonemes
Supra-phonemic information

Question 7

Phonemes

Answer 7

Buildings blocks of the vocabulary
-Smallest units in the signal allowing meaning distinction.
e.g. “bat” and “mat” have three phonemes and differ by the first one.
-In limited number, hence words are created by combining them in an unlimited number of ways, specific to the language.
-English has 20 vowels and 24 consonants.

Question 8

Supra-phonemic information

Answer 8

-This is the prosody/music of speech (i.e., rhythm, melody and energy)
Examples:
-Lexical stress/accentuation
e.g. TRUSty/trusTEE; ADmiral/admiRAtion
-Tones. In some languages, the same string of phonemes can have a different meaning depending on its pitch contour.
e.g. mă (horse), mā (mother), má (hemp), mà (scold), in Mandarin.

Question 9

Categorical perception

Answer 9

-If we are presented with exemplars along a continuum of syllables between two end-points (e.g., gi-ki), we perceive a whole section of the continuum as one category (e.g., as ‘gi’), and the other section as another category (as ‘ki’), despite the physical changes within the category. In other words, there is a step-like shift indicating a category boundary at some point in the continuum (here between Steps 4 and 5). We experience the stimulus as either one or the other, but not as an in-between.

Question 10

How do we acquire the phonemic categories of the native language?

Answer 10

-Examined the ability of English infants to discriminate non-native (i.e., Hindi and Salish) contrasts during the first year of their life.
-Cross-sectional and longitudinal approaches using the conditioned-head turn paradigm.
-Newborns seem to come to life equipped to deal with any possible phonetic contrast.
-With exposure to the language, non-native contrasts disappear, but native contrasts are maintained!