Acoustics final ch. 6 and 7

dimensions of the vocal folds

the amplitude of vibration (determined by the source)

6 dB more

time the vocal folds are apart

time vocal folds are together

losing amplitude for the harmonics as we go up in frequency (women tend to have more shallow slope than men)

fundamental frequencies

women

yes

women

• formant frequencies may be more difficult to estimate
• changing the analysis bandwidth is often helpful
• do this by changing the number of points in the FFT

• women's are 6 dB stronger
• women lose 6 dB per octave, men lose 12 dB per octave in slope

integer multiples of the fundamental frequency

harmonics are further spread apart for women

smaller vocal tracts

• rise-fall
• flat
• fall

• harmonic doubling
• biphonation
• vocal tremor
• noise components
• nasalization

the changing length of the vocal tract

• at puberty
• stability from 4 to 24 months
• lowering from 25 to 36 months
• the shape as well as the length of the vocal tract changes

• period perturbations
• timing of the vocal folds coming together gets messed up

• amplitude perturbations
• the distance of the vocal folds traveling apart is not relatively the same

a small change

no

• more air than voice
• little amplitude, vocal folds barely coming together

• mean
• min
• max
• range
• standard deviation- most important

just the source (nothing else)

slope drops off quick

• velopharyngeal incompetence (can't control the velum)
• increase in formant bandwidths
• decrease in overall vowel energy (mucus)
• introduction on nasal formant
• rise in F1 and lowering of F2 and F3
• presence of antiformants

air is coming through the nose when it shouldn't be

• motor speech disorders due to neurological damage
• speech is very unintelligible

• rate changes
• articulatory adjustments are neglected
• diminished acoustic contrasts
• incomplete stop closures
• generally, timing and sequencing are interrupted

• if therapy is succeeding, positive acoustic changes should be seen
• if therapy is not succeeding, no changes will be observed
• the spectrogram, however, must be properly interpreted

above

• on the segmental line
• we write above the segmental line about intensity, etc (suprasegmentals)

• speech sounds are not produced in isolation, but in context (syllable, words and phrases)
• english is highly anticipatory
• coarticulators are saving time

• nasal sounds can make vowels nasalized
• rounded sounds can make sounds rounded that aren't normally

• anticipatory
• retentive

• features of a sound appear earlier than the sound
• forward coarticulation (a sound effects the sound before it)

• features of a sound carry over to the next one
• backward coarticulation (a sound effects the sound after it)
• usually has to do with nasality

• phonemes become shorter when syllable length increases (if the word is short, we hold the vowel longer)
• vowel duration gets shorter with more sounds before and after the vowel (trade-off between time and number of units within a word)
• speech rate becomes a primary consideration
• acoustic cues are not tightly bound to traditional phonemes

• speech produced with an effort to be highly intelligble
• contrast with conversational speech
• broadcasters and air traffic controllers
• we use clear speech for people with dysarthria (they need to overshoot to make their speech sound normal)
• overarticulating can make it easier to be heard

• slower (durations longer for phonemes than expected)
• avoidance of articulatory modifications
• greater intensity of consonants (appears darker on a spectrogram)

yes, but less natural sounding

• greater fundamental frequency variability (greater variability in pitch when we articulate more)
• precise timing (not rushing through the words, more equal timing for each word)

• melody of speech
• not confined to phonemes
• observed over much larger intervals
• easier to pick up prosody if you have a larger sample of speech

• fundamental frequency (observed as pitch)
• intensity (perceived as loudness)
• duration (perveived as length)

• patterns of pitch rise and fall plus stress
• content/meaning of message is changed
• stress has more intesity and higher duration on a spectrogram

• giving prominence to a word or phrase
• linguistic stress
• lexical stress
• syllabic stress
• contrastive stress (constrasts with previous info)

• mark the ends of language unites
• pauses
• changes in duration
• adjustments of pitch
• meter (rhythm)- pattern of stressed and unstressed syllables

• gives us a lot of communication:
• auditory segmentation
• affect
• personal information
• speech rate (pauses and vowels are shortened proprtionately more, stress patterns may also change)
• vocal effort (adaptive adjustment to distance, not the same as loudness, neither is the same as intensity)

• clear speech
• prosody
• phrasal stress/emphasis
• boundary cues