Speech and Hearing Science - Instrumentation Chapter 12

With the addition of relevant sections from chapter 4, 6, 7, 8

the main objective of any measurement is to

to minimize the measurement error

this involves acoustic measurement, aerodynamic measurement, and physiologic measurement

factors to be aware of during measurement

what is to be measure

• determine what you are measuring

appropriate instrumentation of measurement

• with what are we using to measure

calibration of instrument to ensure correct measurement

• does it work correctly

clear and specific instructions for procedure

• how/when/what to speak

it is important to be mindful that potential ___ of ___ may be beyond our control

potential sources of error

instruments use transducer. what’s that?

transduction! It is the process of changing one energy to another.

ex: microphone captures acoustic energy and convert it to electrical energy

the goal is to obtain a simulacrum as close as possible to the original signal

sensors for capturing speech

pressure sensor (microphone)

airflow sensor (pneumotachometer)

electrical activity sensor (EMG, EGG, ECG, EEG, ERP)

force, position, light, contact, ultrasound sensor

acoustic signal capture/processsing includes a microphone, explain the parts

microphone type - could be handheld, lavalier, or headset

• head set is preferred. for handheld the distance might not be constant. friction noise may be present for handheld.

transducer type - could be dynamic (inductance) , or a condenser (capacitance)

• condenser - laboratory setting. ex:electric microphone

• dynamic - external noise is present

directionality - could be an omnidirectional, or cardioid. this would have the proximity effect which influences the flat frequency response

• does the microphone pick up what’s right in front of it or picks up the full 360 degrees.

• omni in laboratory setting. no external noise. 360

• cardioid = picks up one direction. may not always be a flat frequency though!

connectors

xlr (for quality microphones) or trs

connects signal with comp

recording environment is important. ambient noise (measure by sound level meter) electromagnetic interference ( measure by oscilloscope)

sound treated rooms may not always be required! there are controlled sources of nosied makers (turning off fan, a.c, closing door)

frequency response for signal capturing. what is a good range of frequency fluctuation?

if its less than 2 dB fluctation across the frequency that is acceptable!

how much intensity level is required for the device to be trigger?

that is referred to as sensitivty

dynamic range is (max SPL to self noise)

***both metrics are important!

amplification

preamplifer (gain properities, freq response, dyna ic range)

the key is that the input is to the next level is a true representation of the imput of what the preamplier recieves

digitization

an (analog) signal converted to be analyzed (or digitized)

to be mindful

• sampling frequency (4100 Hz in Praat) this should be twice the max frequency . this is Nyquist-Shannon theorem.

quantizations

• how many steps (or bits) in a cycle will be consider/ will be lose? the higher the better. b = bits. 2^b

• erros are called quantization error, peak clipping //digital analog conversion (DAC)

datat storage referring to acoustic signal capturing

compressed - (used to be a necessity) - mp3, wma

uncompressed (better) - .wav and .aiff

*remember the position of microphone is important. it should be 45 degrees from breath stream, 3-4 cm from mouth. this is important because if you need to replicate it , everything should be exact

eletromyography

measures motor unit action potentials

electrical sensors

its surface. Includes needle and hooked-wire electrodes

provides patient feedback during therapy

inductance plethysmography

measures changes in lung volume during breathing (and other relevant contributions of the abdomen and ribcage)

this involves chest wall kinematics

includes a band around chest

instrumentation for vocal folds

glottopgrahy

stroboscopy

high speed laryngelal imahing

glottography

analyses of vibratory movement of VF during phonation

glottal opening focus

stroboscopy

provides stimulate of VF vibration based on images with strobe light flickering at rate slower than rate of vibration

1 pic = 1 cycle of vibration

various stages of flickering to prevent pictures of exact same point in frequency

high speed laryngeal imaging

high speed photography

images at higher rates

good for irregular vibration cycles! (high jitter value)

videokymography

focuses at a midline of vibration of VF, what is happening.

high speed imaging

photoglottography

amount of light present through an endoscope light source passing through the vocal folds - measured by a light sensor placed externally on the neck just below the vocal folds’ level

electroglottography (EGG)

common for SLP

pair of surface electrodes on the thyroid lamina, how ell does the high freq low voltage current flow from one electrode to the other.

depending on how the current travels across VF we will get a specific measurement

EGG open phase vs closing phase

open - open to entire duration ration

speed - open to close

contact - contact to closed during entire duration ratio

sound spectrogrpahy

wide band - formants (praat)

narrow band - harmonics

spectral analysis

distribution of energy across frequencies at a given moment. plot frequencies on one axis and intensity at another axis.

Linear Perdictive Coding - formants and Fast Fourier Transformation harmonics

microphone is a

sensor and transfusers

all sensors are

transfusers

all transfusers are

not all sensors!

long term average spectrum (LTASS)

spectrum of energy over a window of vowel pronation, not just a specific point in time

another analysis of acoustics

harmonic to noise ration

harmonic energy needs to be higher than noise when using periodic signals.

cepstral measurement - good for aperiodic sources

• ceptral peak performance - relative amplitude of the dominant cepstral peak

inverse filtering

information on glottal airflow is extracted by removing the vocal tract filter characteristics. estimation of glottal flow from the speech signal)

these measures use recorded speech to analyze phonation

vocal tract imaging

xrays!

computed tomography CT - uses narrow xray beam, distinguished tissues with similar density

MRI- posture of vocal tract. fMRI-functional neural imaging

ultrasonography (SLP fav) high freq sound waves particularly tongue mvtm

aerogynamics and articulation

pneumotachography - airflow msrmt , face mask, nasometer, spirometer

hypodermic needle (through CT membrane - connected to pressure sensor. lung pressure measure directly

intra-oral pressure transducer ( saying /p/ multiple times airway interruption method. pt speaks against a presusure it measured airflow that way)

spirometer what it looks like

articulators measured with

point trackers

xray microbeam with up to 100 pelletes positions per sec

electromagnetic midsagittal articulography

induced currents are measures. led sensors

OR

beams attached to strain gauges.

beams attached to rigid wired connected to lips/jaw

electropalatography

contact btwn tongue and soft palette

very important

touch sensitive electrodes embedded in a thin acrylic plate

limitation of intrusive methods= perturbation