final exam new info Flashcards
cognate
when a sound has the same manner and place of production but differs in voicing
-voiceless cognates are produced with more muscle tension and greater airflow
what frequencies do voicing cues occur in
lower frequencies
what frequencies do acoustic cues occur in
middle frequencies
what frequencies do place cues occur in
high frequencies
voiced vs. voiceless on a spectrogram
-longer closure duration for voiceless and a shorter duration for voiced
-aspiration cue within voiceless sounds (striation in higher frequencies)
stop’s on a spectrogram
no sound
fricative’s on a spectrogram
noise striations will be seen
affricate’s on a spectrogram
gap followed by noise
-think of it as a stop then a fricative
sonorants (nasals, liquids and glides) on a spectrogram
appears like a vowel with the formants but with varying voicing along the way
sonorant vs. stop
the stop will have no noise whereas sonorants will have noise
sonorant vs. fricative
sonorants have formant noise where as fricative’s are constant vertical noise
sonorant vs. affricate
sonorant is formant like noise whereas affricates are a stop then vertical appearing noise
voiced vs. voiceless stop on a spectrogram
longer stop gap for voiceless then the voiced
stop vs. fricative on a spectrogram
a stop will have a noise gap and the fricative is constant vertical noise
fricative vs. nasal on a spectrogram
fricative is constant vertical noise and a nasal is a formant appearing sound
how does redundancy aid in speech recognition
it adds in extra assistance for understanding what is being said
intrinsic redundancy
present within the listeners auditory system
-multi sensory processing, binaural processing and can reduce the function effects of lesions
extrinsic redundancy
present within the speech signal
-having multiple acoustic cues such are voice, manner and place
-probability
how does probability relate to speech perception
as the listener hear sounds or words in succession, each is constrained by context
suprasegmental feature
linguistic features
6 suprasegmental aspects
-temporal cues : speaking rate and rhythm
-frequency cues : pitch and intonation
-amplitude cues : stress and juncture
how is emotion conveyed acoustically
due to changes in various features of the talkers voice
why is emotion perception important in children
social cues, being able to communicate effectively with their peers
-development of emotions
why is emotion perception important in adults
provides the context of how the person is feeling and wanting to response appropriately
-relationships and communication abilities
how is emotion perception affected in older adults
they have an increased difficulty in emotions in general and hearing loss impacts that
how is emotion perception affected in CI users
-vocal emotion recognition is impaired for adults with severe HL who use CIs
-CI speech representation has a narrower intensity range
-poor representation of speech
how is emotion perception affected in children
-children with severe HL can exhibit emotional competence
-accuracy of emotion perception in children with moderate to profound HL is lower than that of NH children
short term characteristics of speech
of a single talker, varies from moment to moment
temporal envelope
traces peaks and valleys of the signal
when is temporal envelope an important cue
gives syllables their cue
temporal fine structure
fast changes in amplitude within the waveform
when is temporal fine structure an important cue
pitch and melody cues
modulation depth
variability within amplitude as seen within speech signals
-the more depth there is, the clearer the sound is
why is modulation depth important for speech
it gives speech clarity
-you get a less robust signal to perceive what is being said when this is affected by hearing loss
how is modulation depth affected by background noise
the noise can fill the gaps of the depth and therefore create a reduced ability to detect what is being said
long term characteristics of speech
average over time with multiple talkers
long term average speech spectrum (LTASS)
the amount of sound energy at each frequency for an average of multiple talkers or an average of time period
-frequency on x, level on y
what is the average levels for speech
soft - 40
average - 50
loud - 70
what is the RETSPL for speech
supra-aurals - 20 dB
inserts - 13 dB
speech intelligibility index
the percent of speech that is available to a listener
-uses frequency weighting
what is the SII chart based off of
listeners with typical hearing thresholds in quiet
what two questions are addressed by the SII
-are all speech frequencies equally important for speech recognition
-does the entire speech signal need to be audible for maximum speech recognition
what frequency bands are the most important for speech discrimination
2000-3000 Hz
