Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Applied Speech Science > Lecture 2, 3, 4 > Flashcards

Lecture 2, 3, 4 Flashcards

You may prefer our related Brainscape-certified flashcards:
Biology 101 flashcards
1
Q

What is a spectrum?

A

A spectrum is a visual display of a complex sound wave that shows the amplitude of the frequency components at one point in time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the horizontal and vertical axes of a spectrum?

A

Horizontal axis: Frequency in Hz

Vertical axis: amplitude in dB-SPL

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is a spectrogram?

A

A visual display of how the spectrum changes over time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What are the horizontal and vertical axes of a spectrogram?

A

Horizontal axis: Time
Vertical axis: Frequency
*Amplitude is represented by the darkness of the display

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is a filter? What is its purpose?

A

A device that restricts the range of frequencies that are present within a sound. They increase the amplitude of desired frequencies and decrease the amplitude of undesired frequencies.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is a low pass filter?

A

A low pass filter only transfers sound energy at frequencies BELOW a certain cutoff

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is a high pass filter?

A

A high pass filter only transfers sound energy at frequencies ABOVE a certain cutoff

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is a band pass filter?

A

A band pass filter only transfers sound energy WITHIN a certain band of frequencies

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the response curve?

A

The slope of the cutoffs in a band filter: the cutoffs are NOT discrete. Rather, the amplitudes of frequencies around the cutoff range get smaller and smaller. Therefore, not all frequencies in the band pass filter are represented equally

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

How is a narrow band spectrogram useful?

A

A narrow band spectrogram is basically a band pass filter with a very SMALL range of frequencies (i.e. 30-50). With a smaller bandwidth, a greater amount of frequency detail can be seen. That is, with an active bandwidth, we know that certain frequencies within the bandwidth are present. Therefore, we are able to tell (with greater precision) which specific frequencies are present.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is a limitation of a narrow band spectrogram?

A

The window of analysis needed to observe frequencies in this amount of detail is relatively LONG (25 to 30ms). Therefore, narrowband spectrograms have POOR TIME RESOLUTION. Therefore, rapid changes that occur in the waveform (shorter than 25 to 30ms) are not resolved, because they are shorter than the window of analysis itself.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

When should the narrow band spectrogram be used?

A

When we want to know detailed information about the harmonic structure of a sound signal: for example, when analysing the periodicity of an elongated vowel

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How is a wide band spectrogram useful?

A

Wideband spectrograms have good temporal resolution. The way that sound energy changes across frequency can be observed over much shorter periods of time (i.e. down to 3-5ms).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is a limitation of a wide band spectrogram?

A

Wideband spectrograms have poor frequency resolution. We only see what happens in steps of 3 to 5 ms (which equates to steps of 300 – 500Hz).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

When should a wide band spectrogram be used?

A

Wide band spectrograms are useful for observing rapidly changing events, like glottal pulsing. They also tend to show vowel formants more clearly, which can be useful when examining the RESONANCE of the vocal tract

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What do we call the vertical stripes aligned with each glottal cycle? Where can you observe these?

A

Striations; on wideband spectrograms

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What does the fundamental frequency physically represent?

A

The rate or frequency that the vocal cords open and close during phonation (also called glottal pulsing)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Describe the source filter theory of speech production

A
  1. Sound is produced at the larynx (source)
  2. The vocal tract acts as a resonator – it amplifies the harmonics that match the resonant frequency of the vocal tract at that time.
  3. The size of the vocal tract determines its resonant frequency
  4. The point of constriction determines the size of the vocal tract – and therefore acts as a FILTER that amplifies certain frequencies but not others.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What are the two main components of the source filter theory of speech production?

A
  1. The source (sound generated at the larynx)

2. The filter – the shape of the vocal tract modifies the sound by amplifying certain frequencies (formants)

20
Q

How do we manipulate the filter?

A

By changing the shape of the vocal tract

21
Q

What happens when we manipulate the filter?

A

We change the transfer function or the resonance characteristics of the vocal tract

22
Q

What are formants?

A

The frequencies that are passed through the vocal tract with greater amplitude (because they match the characteristic resonances of the vocal tract).

23
Q

What is the difference between harmonics and formants?

A

Harmonics are a feature of the SOURCE spectrum. Formants are those frequencies that match the resonant frequency of the VOCAL TRACT

24
Q

How do we produce different sounds according to the source filter theory?

A

By manipulating the filter (i.e. vocal tract shape) we change the resonant frequencies (and therefore change the particular frequencies that are amplified).

25
Q

What do the formants look like in central vowels?

A

The vocal tract has a uniform area across its length in central vowels. Therefore it is most like a closed end tube, and the higher resonance frequencies (aka formants) are at every odd multiple of F0. They are therefore EVENLY SPACED.

26
Q

What do the formants look like in front vowels?

A

Front vowels have a wide separation between Formants 1 and 2, and a small separation between Formants 2 and 3

27
Q

What do the formants look like in back vowels?

A

Back vowels have a small separation between Formants 1 and 2 and a wide separation between Formants 2 and 3

28
Q

How is Formant 1 related to tongue position?

A

Formant 1 is inversely proportional to tongue height. Low Formant 1 frequencies are associated with a high tongue position, and high Formant 1 frequencies are associated with a low tongue position.

29
Q

How is Formant 2 related to tongue position?

A

Formant 2 is proportional to tongue advancement. Front vowels have high Formant 2 frequencies; back vowels have low Formant 2 frequencies; central vowels are in between. See Figures 2 and 3 for examples.

30
Q

What is the difference between cardinal vowels and diphthongs?

A

Diphthongs involve a shift in vocal tract position, while cardinal vowels have a single target position. Therefore, diphthongs have formant transitions, and cardinal vowels do not.

31
Q

Which acoustic features are characteristic of stop consonants?

A
  1. The stop gap interval

2. The stop release

32
Q

What does the stop gap interval represent?

A

The period where the oral cavity is completely closed (and therefore no sound is released). During closure air pressure is built up within the oral cavity.

33
Q

What does the stop release represent?

A

The rapid opening of the oral cavity, resulting in the release of air and a corresponding burst of noise. This noise is called frication. The air rushes through a small gap, creating turbulence.

34
Q

What happens when a stop consonant is followed by a vowel?

A

The articulators are still moving from the closed position held during the stop gap to the open position required for the particular vowel

35
Q

What do the formants look like in a stop + vowel sound?

A

There are rapid, steep formant transitions, reflecting the rapid change in vocal tract shape.

36
Q

What is anticipatory co-articulation?

A

The way that our articulatory gestures of phonemes also incorporate the movement requirements of the following sounds. Therefore, our speech motor plans are likely to exceed single phonemes.

37
Q

What are three acoustic features that distinguish voiced stops from voiceless stops?

A
  1. Aspiration (a puff of air) – in voiceless but not voiced
  2. The presence of a “voice bar” – a low-frequency, low-amplitude formant-like band – in voiced but not voiceless
  3. VOT is longer for voiced than voiceless stops
38
Q

How are fricatives produced?

A

Air rushes through a very tight (though incomplete) constriction, creating APERIODIC turbulent noise

39
Q

Acoustically, how can we tell the difference between fricatives produced inside the mouth (e.g. s) and fricatives produced at the front of the mouth (e.g. f)

A

Fricative sounds produced inside the mouth have relatively CLEAR SPECTRAL PEAKS.
Fricatives produced at the front of the mouth have fairly flat spectrums (i.e. sound is distributed MORE EVENLY to a range of frequencies in fricatives produced at the front of the mouth.

40
Q

What causes the difference in the spectral peaks of anterior and posterior fricatives?

A

The presence or absence of RESONANCE within the cavity of the vocal tract. Resonance is present in fricatives with a more posterior place of constriction.

41
Q

Acoustically, how do alveolar (s, z) and post alveolar (sh, dz) sibilants differ?

A

The frequency of alveolar sibilants is much HIGHER than post-alveolar sibilants. This is because the constriction is further forward in the mouth, resulting in a SMALLER cavity through which the sound travels BEFORE leaving the mouth.

42
Q

How are anterior fricatives (e.g. f) represented acoustically?

A

By very high resonant frequencies (because the cavity beyond the point of constriction is basically non-existent

43
Q

How are affricates represented acoustically?

A

Place of articulation is marked by the starting frequency of the second formant transition in the vowel (as in stop consonants) as well as the location of spectral peaks (like fricatives)

44
Q

How does intensity differ from nasal sounds compared to vowels (in terms of frequency and intensity)?

A

Nasal sounds generally have a lower intensity, because the nasal cavity has greater sound absorption properties.
Nasal sounds have lower frequencies because the sound travels through a longer cavity.

45
Q

How does F2 frequency differ across nasal sounds?

A

F2 increases as the place of articulation moves back in the mouth. F2 is highest in /ng/, then /n/ and then /m/

46
Q

What is a general relationship between the place of constriction in consonants and F2 frequency? To which types of consonants does this rule apply?

A

Anterior places of constriction = lower F2
Posterior places of constriction = higher F2
This rule applies to stops, nasals, and glides

Applied Speech Science (4 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions