Docsity
Docsity

Prepare for your exams
Prepare for your exams

Study with the several resources on Docsity


Earn points to download
Earn points to download

Earn points by helping other students or get them with a premium plan


Guidelines and tips
Guidelines and tips

COMD 5070 Final Exam-with 100% verified solutions 2024-2025, Exams of Advanced Education

COMD 5070 Final Exam-with 100% verified solutions 2024-2025

Typology: Exams

2023/2024

Available from 08/11/2024

studycamp
studycamp 🇺🇸

3.9

(29)

3.2K documents

Partial preview of the text

Download COMD 5070 Final Exam-with 100% verified solutions 2024-2025 and more Exams Advanced Education in PDF only on Docsity!

COMD 5070 Final Exam-with 100% verified solutions

2024-

neither invariant nor linear The speech signal is perceptual normalization When a listener simplifies the processing of the speech signal to smooth out superfluous variability, it is called categorical perception One type of quantal perception that occurs as we listen to speech is called syllable components are presented separately, but simultaneously, to each ear In duplex perception: the place and/or manner of articulation is perceived to change In the McGurk effect: using contextual, linguistic, and cognitive cues Top-down processing in the perception of speech involves motor theory Which speech perception theory has been criticized as being so vague that it is more of a philosophy than a theory or model? perceptual magnets enhance differences between distinct phonemes

Cohort theory based on similarity of acoustic features of the beginning of words dominant language exposure gradually decreases discrimination of non- native speech features research with infants has shown that a refinement of everyday thinking What is science? empirical, deterministic, predictive, parsimonious What are the 4 features of the scientific method? semitone A nonlinear increase in frequency as you go from low to high 12 How many semitones are in one octave? 5.9% There are ________ more semitones in the octave before the last Sampling rate the frequency with which numbers are stored or written down to represent the analog signal that you're sampling.

the higher the sampling rate, the better the playback quality How does sampling rate relate to playback quality? Nyquist frequency The highest frequency the can be reproduced- or the highest sampling rate high pass filter allows high frequencies through and holds back lower frequencies low pass filter blocks high frequencies. The best way to avoid aliasing during a digital recording. band pass filter allows a band of frequencies though and holds back both higher and lower frequencies band reject filter holds back a band of frequencies and allows both higher and lower frequencies through frequency domain a line spectrum shows the frequency components of a period sound

time domain a waveform represents sound directly, air pressure changes over time FFT spectrum -clearly shows harmonic energy -each peak is a harmonic -less clear at showing formants -more revealing of source LPC Spectrum -shows spectral envelope -good at revealing formants -does not show harmonics -more revealing of filter time, frequency, intensity In a 3D spectogram, the x-axis represents _________, the y-axis represents ____________, and the darkness represents ________________

vocal modulation rhythmic changes that are predictable in fundamental frequency and in amplitude in the voice; gradual increases and decreases that occur over the course of many glottal cycles perturbation imperfections in the vocal signal tracheal puncture, esophageal pressure/balloon What are the direct measures of subglottal pressure? tracheal puncture -make a hole in the trachea below the larynx and put a pressure transducer in the trachea -not for the faint of heart -give direct, accurate measures during speech Esophageal pressure/balloon -swallow sensor partway into esophagus -measures the pressure on the shared wall of the trachea and esophagus

-pressure is lower than lung pressure -not very practical or common measure intraoral pressure at a particular time (during /p/ closure) How do we estimate subglotta pressure? voice intensity changes What happens if we change subglottal pressure? volume divided by time How do we calculate average air flow? Pressure=Flow x Resistance How do pressure, air flow and resistance relate? Ohm's law Pressure drops across the glottis, for a given pressure higher resistance = lower flow and lower resistance = higher flow. pressure and resistance and flow are all linearly related resistance (how tightly we adduct the vocal folds) and pressure (how hard we pump air with lungs) What determines how much air flows through the larynx?

measure flow, estimate Psub and calculate estimated laryngeal resistance by dividing Psub by flow How can we compute an estimate of laryngeal resistance? source-filter theory Larynx is the sound source, the output is sent into the vocal tract. The vocal tract is the filter. In theory, the two systems should be able to operate independent of each other, but there are circumstances under which they interact loudness (less lung pressure to speak softly), pitch (stretch the vocal folds to raise frequency), voice quality (adjust space between vocal folds--larger space=breathy voice) What are examples of a change to the source? differences in vocal tract tube affect resonant frequencies, specific features change depending on constriction locations, movement of tongue and other articulators What are examples of a change to the filter? increase as you lower the jaw and tongue, you _____________ the F1 frequency high front vowel Which location of vowels has the lowest F1 frequency?

low back vowel Which location of vowels have the highest F1 frequency? up As the tongue move forward, F2 frequency goes ___ stop gap/stop closure, voice bar, transient, frication, aspiration, voice onset time What are the acoustic features of stop consonants? Acoustic characteristics of burst aspiration Burst: on flow signal (waveform), shown as a peak with a drastically dropping pressure afterward on spectogram, shown as a black section (with an abrupt start after stop closure) Aspiration: may follow a voiceless stop -only when stop is followed by a vowel on a waveform: weaker noise signal On a spectogram: weaker region of frication, and mostly in lower frequencies women; EMT polyglot (grew up speaking English, but acquired a second language later on in life)

In recognizing emotional tone in speech, which listener group preformed best between men and women? Which group preformed best overall? Linguistic prosody: pitch-based type of prosody: it's in the mail vs it's in the mail? Linguistic prosody: pause-based type of prosody: a woman without her man is nothing vs a woman: without her, man is nothing affective prosody type of prosody: don't talk to me in that tone of voice! forward/anticipatory coarticulation an earlier sound is influenced by a later sound backward/retentive coarticulation Influence of a later sound by an earlier one the nasal cavity acts as an antiresonance and tends to have a damping effect and can influence the clarity of a vowel that's being produced. Formants are substantially damped How does perceptual severity of hypernasality relate to physiology?

False T/F: perceived nasality is linked rigidly to VP port orifice size high Nasal sounds such as /m/ or /n/ are very _____ in air flow vowels _________ are very oral sounds and therefore very low in nasality pressure consonants nasal air flow is near zero for ___________________-- Kay Elemetrics Nasometer, aerodynamic measures, and OralNasal Mask system What types of instruments quantify nasalance? electromyography, x-ray studies, and x-ray microbeam What types of instruments look at the velum? How acoustic nasometry works conceptually You have two microphones, one is set above a divider plate and one is set below it. This divider plate is placed on the front of the person's face so that it sits above the upper lip and below the nose. The upper microphone

records nasal acoustic signals and the lower records oral acoustic signals. The display shows degree of nasalance (relative amount of nasal energy to total energy) electrical activity measured during a muscle contraction following a neural stimulation What does EMG show us? central nervous system EMG electrical activity represents signals sent to the muscle. These signals originate in the __________________________ and may reveal details of neural control rectification, smoothing, averaging What are the signal processing approaches used to help us simplify complex EMG signals palatometry numeric data produced y a plastic pseudo-palate that has contact electrodes. Any tongue contact is detected by the computer and patterns of contact is displayed assessing articulation problems and provided biofeedback during therapy How is palatometry useful? motor equivalence, coarticulation, does not reveal tongue movements

What are the limitations of palatometry How speakers time respiration and talking even the simplest utterance requires planning and control. Correct timing and depth of inhalation, appropriate respiratory phase for phonation, suitable level of vocal fold adduction, relative timing of voice onsets and offsets, multiple vocal tract structures move for each phoneme, timing and coordination are precise in normal speech models a simulation used to explain or test theories a broader conceptual explanation based on many observations hypotheses a tentative prediction on a specific topic- should be testable acoustic theory of speech specifics of movement control are less important than the acoustic/perceptual result. motor equivalence allows flexibility: there's more than one way to achieve the goal articulatory theory of speech

vocal tract configurations correspond to sounds. Movement patterns are stored for retrieval and use aerodynamic goal theory of speech speech sounds involve regulating pressure and flow. Pressure stability relies on correct valving That it really doesn't matter that much which specific movements are made, as long as the acoustic result is acceptable; backing up the acoustic goal theory What does a bite block show us about speech? speech tends to adapt really well, tongue and lip movements get bigger to compensate How does speech change if you fix the jaw position? no evidence to say it has any value in helping improve speech: speech isn't generally affected by the strength of your tongue what are some problems with the non-speech oral motor exercises approaches? Auditory feedback feedback is essential as babies babble, look for feedback and adapt sounds as they start to mimic adult speech. It helps us learn the association between movements and the sounds that result from those movements. As we become proficient speaker, we use it in a more general sense to maintain the quality: quality control

single processor model suggests that we have one pool of resources in the brain and this has to be shared out among all the competing demands for it switching model suggests that we rapidly alternate between the different tasks that we are doing multiple processors model suggests that we have different pools of resources that we use for different tasks functional distance model is linked to what we know about brain localization (right hemisphere controls left side of body). This model suggests that if you're doing something difficult with your right hand at the same time as you're doing something with the left hemisphere (like language processing) then you're competing for the same territory in the brain motor program theories Plans are made before movements begin, planning involves movement selection & sequencing. Production is like playing out a written musical score. Reaction time data support this theory. open-loop control of skilled movements

in this case, we learn over a number of trials that a certain action that we take results in success or not. If our previous efforts to produce a certain sound end in success, then we can confidently use them again with the same expectation. It's predictive, based on experience. closed-loop feedback we rely on incoming signals to guide our ongoing actions. the adjustments that we make will help us correct and improve the results. Ongoing, leads to corrective adjustments. continuous perceptions It's a continuous function, when graphed it provides a gradual slope without any big gaps or jumps. Example: we can link intensity and the human perceptions of loudness and any adjustments in intensity would result in slight changes in perceived loudness categorical perception Not a gradual change when graphed, but rather an abrupt transition. For example: when graphing people's responses when identifying whether they perceive the /b/ sound or the /p/ sound at different VOT, when the VOT is 20 ms, about half perceive /p/ and half perceive /b/, The physical stimulus has been changed by a small increment each time, however there was no perceptual difference between 0-10 but a large difference between 10-20 meaning listeners tend to respond in a very non-linear way phonemic memory A lasting memory that we gradually acquire throughout our lifetime because of all the many repetitions we hear of each sound in our native language. We gain a strong and lasting memory through repeated exposure to the specific sounds used to communicate with people around us.

long-term sound template phonetic memory can be compared to a ______________________________. We compare incoming sounds against this template. Identification experiments one stimulus at a time, open-response set or forced choice (multiple choice), no right or wrong answers, just response patterns. the person simply has to report by either speaking to the researcher, pressing a button, or writing down the sound that they perceived. discrimination experiments no need to label what was heard, the job of the listener is to determine whether two stimuli were the same or different. Answer can be correct or incorrect heart rates; drops, not different, different, not different Infants are open to more categories and the language categories they are exposed to shape their abilities. Researchers have monitored babies' _______________ as they are exposed to different sounds (/b/ and /p/) when they perceive a different sound, their heart rate ____________slightly. VOT of -10/0 ms was ______________ VOT 20 / 40 ms was ___________ VOT 60 ms / 80 ms was ________________ pattern playback machine

opposite of a spectogram. Scientists draw sound waves on a graph and the machine produces the sounds recorded on the graph. This allows manipulation of isolated components It is easier to perceive small changes when they are close to the phonetic boundary while differences that are distant from the boundary are less clear. We have better sensitivity where it makes a difference How do small acoustic changes result in a change in the identity of a phoneme? (within vs across-category changes in sounds) how we parse words for a stream of sounds Segmentation. We use top-down knowledge. If we have linguistic knowledge of which words are potential candidates to be segmented then we can apply this knowledge to the sound string that we hear. Phonotactics can also be useful to help narrow the options perceptual assimilation When we hear an unfamiliar foreign sound, we typically put it into one of our own categories. McGurk Effect If a researcher records video of a speaker saying "ba ba ba" and then "ga ga ga" and switches the audio and the video of each, the listener will hear what they see the speaker saying, not what the audio is actually saying. perceptual magnet effect acoustic variants close to the prototype are harder to distinguish from it; it pulls them to it. Acoustic variants away from the prototype are easier to

distinguish from each other; there is no magnet effect. Acoustic variants in a foreign language are easier to distinguish; magnet effect relies on experience with a language.