COMD 5070 Final

3D spectrogram - answer-hybrid display (not a time or frequency display) because it is a combination of time and frequency. a) Shows us how the strength and its range of frequencies is changing or evolving over time from left to right along the x-axis. b) Shows frequency changes over time and changes in amplitude of frequency components whether they are weak or strong. 3D spectrogram features. (what are the 3 dimensions) - answer-Time Frequency Intensity 4 features of scientific method - answer-empirical deterministic predictive parsimonious A fourier analysis reveals ___ about a complex wave: - answer-x axis = frequency y axis = amplitude Peaks are individual harmonic components Each column represents a frequency component and it splits individual sounds out of a combined total to show the individual components and relative strength of those components A line spectrum shows...... - answer-the frequency components of a periodic sound. A single vertical line represents a sine wave. acoustic goal theories of speech [what do these theories say about how speech is controlled?] - answer-"the targeted goal is the correct production of speech rather than how that production is accomplished. Acoustic Nasometry (how does it work conceptually) - answer-you have two microphones. One is set above the divider plate, and one below. This divider plate is set on the front of the person's face so it sits above the person's upper lip and below the nose. So one of the microphones picks up energy that comes below the mouth, and the other picks up energy that comes from the nose. The relative proportions of the energy from these two microphones can be represented as nasalance. This is useful in clinical assessments because you can determine if a person falls within normal limits for nasalance for a given utterance. It can also be used to provide feedback during therapy so the person can see how they're managing the oral nasal balance of speech production. Acoustic parameters that may change when a word is stressed: - answer-fundamental frequency intensity duration Across category change - answer-Heard as different (when you span sounds out you can hear change between sounds). Anticipatory (forward) coarticulation - answer-earlier sounds are influenced by a later sound . ex: spoon which causes /s/ to lip round. Any voice without any perturbation will sound______ - answer-artificial Articulograph - answer-It's a very expensive and complex system. But it does represent the gold standard in measuring articulatory movements. You can see a girl inside the system. She has transmitters in front of her. Those white circles, one near her right at hand, at the bottom, and one to the right of our view. These transmitter coils each send out a different frequency. There are three more behind her that send out their individual radio frequencies. Notice coming from the corner of her mouth are small, white wires. These are attached to sensor coils that would be on her tongue, lips, or face. The position of each coil is tracked in a complex way by the computer software by figuring out how strong relatively each frequency is from the six transmitter coils. It does a form of triangulation in space to figure out in three dimensions where each sensor coil is relative to the frame. It does this in real time, continuously updating multiple times per second. You can imagine this system turns out a complex and large data file. But it does allow us to extract from this data stream soup what is going on in terms of the articulators, relative to the skull. This is an expensive system. It's about as expensive as a high end European sports car or a small house. auditory feedback [how does it help with learning to speak vs. how we control speech as adults?] - answer-Feedback is essential for learning (babies experiment with feedback to help them make adaptations so sounds match parents/caregivers). Feedback is also important for quality control (e.g., adventitious deafness is when deafness occurs postlingually and their quality of speech declines because they are lacking the feedback. Patients that then received a cochlear implant improved their quality of speech.) auditory feedback is very important as infants when first learning speech and language. Aud fdback helps us learn association btw movements and their resulting sounds.Once we r proficient speakers it is used more to maintain quality and not moment by moment monitoring. Averaging - answer-Done with repetitions of the same behavior order to get a composite picture. Band pass filter: - answer-allows a band of frequencies thru but removes higher and lower frequencies Band reject filter - answer-allows both higher and lower frequencies thru but attenuates a band of frequencies Categorical perception - answer-In categorical perception, acoustics (the physical stimuli) are varied along a continuum, with gradual changes from start to end. At one point, the perceived sound abruptly changes. This translates to a discontinuous perception of the physical continuum. change of filter in source filter theory - answer-changes depend on constriction sizes and locations in the vocal tract articulators moving changes the constriction areas change of source in source filter theory - answer-Changes: Vary Loudness, Pitch, Voice Quality, & Phonation: Closed loop control of skilled movements - answer-Closed-Loop FB: Ongoing; Leads to corrective adjustments. Rely on incoming signals to guide ongoing actions. Adjustments made help correct/improve results. Ex. Driving along twisty road depends on your view & how car travels to make adjustments to speed/direction in order to stay on road. CONSTANT FB needed to make sure car driving safely. Very different from open-loop where once arrow released there is no further input that archer can give it. Continuous perception - answer-In continuous perception, the physical stimulus is varied along a continuum, with gradual changes from start to end. The perception of these changes is continuous, with no big jumps. coordinating speaking and breathing [how do speakers time respiration and talking relative to each other?] - answer-~We need to be in the right respiratory air and have breathed in enough air in order to produce phonation for speech -We need to have breathed in first before speaking Even the simplest utterance requires planning and control. Timing and coordination are precise in normal speech. There are several factors that we must coordinate: a) Correct timing of respiratory phase (i.e., don't begin speaking at the end of an exhalation) b) Depth of inhalation (i.e., deeper for louder/longer utterances) c) Suitable level of vocal fold adduction (too much = not enough or no oscillation; too little = weak/breathy voice) d) Relative timing of voice onsets and offsets (voiced/voiceless VOTs are timed by milliseconds) e) Multiple vocal tract movements for each phoneme Describe the events that take place during stop consonant burst.... - answer-During Closure: Silence, then Burst. Silence During Closure: Brief instant when no snd leaves VTract & reflected in microphone signal as brief silent event. Sometimes small amt of