CSD 313 Exam 3

if you are seeing this please leave me a 5 star review it is so good for my self esteem

How we model the vocal tract

Can be modeled as a series of an unlimited number of tubes - “Infinite tube model”

Are the formant frequencies same for everyone

no - it is dependent on vocal tract length and resonating cavity size

Tense vowels

produced with greater muscle contraction, produced at the extremes of articulatory poster, with tongue in higher oral cavity, longer vowels

Lax vowels

shorter, less muscle contraction

Characteristics of Diphthongs

two vowels with the same syllabic nuclei, smooth glide from one vowel to the next

A consonant is inserted, ex: coordinate, cooperate (/w/)

What happens when two continuous vowels are each syllabic nucleus

Why is energy lost in the vocal tract

glottal opening, absorbent walls of pharynx and mouth, friction between air particles

Narrowband filter

provides good resolution of the harmonic (frequency) structure of the source signal

Wideband filter

provides good time resolution of the glottal pulses and formant structure (resonators) of the vocal tract

Male spectogram

Female spectrogram

What does CPP tell us

perception of breathiness and abnormal voice qualities

High pass filter

blocks out low frequencies

Low pass filters

blocks out high frequencies

What does aperiodicity means in the CPP

Can be analyzed with aperiodic sources but CPP will be less

Conventional radiography x-ray

noninvasive, low cost, returns 2D images

Computed Axail Tomography (CAT)

3D reconstruction, similar density tissue can be distinguished, too slow for speech, higher x-ray dosage than the conventional x-ray

Magnetic Resonance Imaging (MRI)

3D reconstruction, dynamic MRI enables us to record with temporal resolution, lower special resolution than CT, must remain completely still, not comfy, noisy

Ultrasound

beneficial for studying tongue movement in speech, identifying tongue contour with clarity, inability to image the tongue tip, large number of data points

Bandpass filter

Type of filter that a vocal tract behaves like

Burst

F1

F2

F3

Stop gap

Aspiration

Properties of fricatives

narrow constriction but not complete occlusion, always sufficient to produce turbulent flow

Properties of sibilants

Greater constriction that is more posterior in oral cavity

Approximates

Glides and liquids

Glides

have a nearly periodic sound source as in vowels, acoustic evidence for place of articulation is formant transition of the neighboring vowel

When is a pseudo palate used?

an array of touch sensitive electrodes embedded in a thin acrylic palate, like a dental retainer - can give unique information about articulation

Voiceless fricatives

frication noise is the sole source

Voiced fricatives

friction noise and phonatory source

Stops formant transition

Slow formant transition in comparison to stops

Fast formant transition

Fast formant transition in comparison to diphthongs

When does the dark /l/ typically happen

after the vowel within a syllable ex: eel, pal, tall, school

Light /r/

is produced with greater tongue advancement

No

Is “road” classified as a light /r/

Low

Is consonant energy high or low

Properties of nasal production

All nasals are going to be voiced; occlude oral cavity and open velopharyngeal port

Formant structure can be syllabic like vowels, but have significant constriction

Nasal murmur = very low F1, F2 and F3 vary

What does Intra-oral Air pressure depend on

degree of constriction of the phoneme and intensity

Ways that hypernasality decreases intelligibility

Introducing antiformants and dampening acoustic energy

Introduction of noise from turbulent nasal airflow emissioins

Decreasing intra-oral air pressure, thereby, decreasing clarity of consonant production

acoustic targets, articulatory gestures, aerodynamic pressures

Targets for “output target” models

Is acoustic invariance a speech perception theory

Yes - there are invariant acoustic patterns in the speech signal corresponding to phonetic features, which remain invariant across speakers phonetic contexts and languages

True

According to motor theory, the listener accesses their own knowledge of speech production in perception

True

The connectionist model is nonlinear and contains a nonhierachical set of components

Feedforward model

signals are used to make articulatory adjustment online; would be used to initiate rapidly sequences skilled movements

Acoustic invariance theory

Acoustic landmarks are important characteristics in which of the following speech theories

Identify the spatiotemporals models

How the articulators move in space and time - refers to the path of an articulator and the timing of the sequence of positions

Criticisms of motor theory

little empirical evidence to support theory, specfic mechanisms of analysis-by-synthesis have not been clearly described, acoustic features vary greatly

Bottom up

processing works in the absence of the knowledge; receive auditory information, convert it into a neural signal, and process the phonetic feature information

Top down

processing works with knowledge a listener has about a language, context, experience; use stored information about language and the world to make sense of the speech

Characteristics of active speech perception

emphasize the cognitive control in perception; including the formation about the phonetic or linguistic interpretation of the information in the acoustic signal

Categories of speech perception

Active vs passive

Bottom up vs top down

Autonomous vs interactive

Aerodynamic targets

Limitations in acoustic feedback in a population with hearing impairment supports which of the following models of speech production?