7 | Perception for (of?) language

Question 1

Hemispherical specialisation

Answer 1

the left brain is usually more adept at recognizing language, this is linked to right ear advantage. Tested in dichotic listening experiments, which used competing sequences of words presented through headphones

Question 2

Diphone

Answer 2

A sequence of two sounds, capturing the important transitions from one sound to the next. Used to create more natural sounding speech synthesis.

Question 3

Variability

Answer 3

The details of any given input may vary. Example, writing styles. This is also featured in speech

Question 4

Exemplars

Answer 4

Multiple stored representations, to account for variability.

Question 5

McGurk effect

Answer 5

The visual cues impacts the audio we hear. “Ba ba”, “fa fa”, und so weiter, und so weiter.

Question 6

Puff of air

Answer 6

A puff of air might make us interpret a sound as a plosive instead of the non-plosive alternative. “B” turns to “p”. This works any way the puff of air is targeted, be it the ankle or the face.

Question 7

Cue trading

Answer 7

If one cue is unclear, such as the plosive part of “p”, then the listener assigns greater value to other clues, such as the relative duration of the preceding vowel

Question 8

Signal continuity

Answer 8

We are better at following a signal from one continous source rather than from different sources. Lies behind the “Cocktail party effect” in which we are able to follow one conversation in a room full of conversations

Question 9

Word superiority effect

Answer 9

Letters are recognized more rapidly in real words than in non-words, or in jumbled letter strings.
We fixate longer on a word while reading than we actually need to recognize the word. If we jump ahead to far in a sentence while reading the meaning of words will be mixed and lose meaning.

Question 10

Deep and shallow ortography

Answer 10

Basically how “wrong” a pronounciation can be vs how it is spelled. English, deep ortography, french or italian, shallow ortography

Question 11

Categorical perception

Answer 11

We will place sounds heard in categories X and Y rather than X-ish or Y-ish

Question 12

Ganong effect

Answer 12

We are more likely to insert a sound which makes sense rather than a sound that does not.

/?esk/ will more often be connected to /desk/ rather than /tesk/