Brainscape's Knowledge GenomeTM

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


See full index

Amplification II > 4 - Compression > Flashcards

4 - Compression Flashcards

1
Q

What was used before compression was possible?

A

Peak clipping

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

How did peak clipping work?

A
  • Peak clipping results in distortion, leads to poor speech perception, and degrades intelligibility at high inputs
  • If the output is no longer a sine wave, as there is missing information to process from in the input signal
  • If anything came over the top amplitude, it was cut off
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the main goal of compression?

A

A main goal of compression is to compress the range of sound levels in the environment so that this range can fit in the reduced dynamic range of the person with hearing loss

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

Why is compression important for SNHL?

A
  • SNHL has a reduced DR (dependent on the slope and configuration of the hearing loss)
  • SNHL needs more audibility to hear the soft sounds without overamplifying the loud sounds
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Why can compression be difficult?

A

Difficult to restore audibility for soft sounds without loud sounds being over-amplified and getting uncomfortably loud

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

What are the 4 main purposes of compression?

A

1) Maintains a perception of “soft”, “moderate”, and “loud” sounds by fitting the sound into the residual dynamic range
2) Maintains a comfortable listening levels for patients
3) Reduces the adverse effects of loud inputs
4) Compression helps to preserve the waveform and helps to maintain speech recognition ability

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

What are the 4 components of compression in real life?

A

1) A loud talker (or the patient’s own voice) may trigger the compression parameters
2) The next talker is a soft spoken person (the hearing aid is in a quieter state)
3) The HA needs to provide more gain for this soft spoken person but it is slow acting, so it provides insufficient amplification for the beginning of this person’s conversational turn
4) Meanwhile, other environmental noises are occurring

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

Compression is what 3 things?

A

Compression is dynamic, changing, and variable

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

How does compression effect the waveform?

A
  • Through temporal pattern changes
  • With no compression, peaks and valleys are well defined (a well defined envelope)
  • With compression, the temporal pattern is altered and the waveform is modified
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Define compression threshold (CT)

A

Level of the signal that will activate compression
Also called “Compression Kneepoint” or “Threshold Kneepoint”

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

Define compression ratio (CR)

A
  • Related to the slope on the INPUT-OUTPUT (I-O) function or degree to which the signal is compressed (how much compression is applied)
  • Example: 10 dB input signal, 10 dB output signal: 1:1, vs.10 dB input signal, 5 dB output signal: 2:1
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Define WDRC

A

WDRC occurs when we are applying compression to a wide range of inputs and frequencies over the speech spectrum (this means a relatively low Compression Threshold (typically below 55-50 dB)

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

Does everyone need WDRC?

A
  • No, not everyone needs WDRC
  • Mild losses don’t need WDRC is they have good DR
  • CHL should have a large DR as long as we can make sound loud enough
  • Mostly just SNHL needs WDRC
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Remember, ____ sounds more natural than ____

A

linear amplification, compressed applied sounds

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

Should compression be used for mild losses?

A

compression may not be needed for audibility and comfort

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

Should compression be used for CHL?

A

May not need compression if the dynamic range is maintained

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

Should compression be used for sloping or moderate/severe loss?

A

Compression may be required at some frequencies, but not others. Intelligibility for soft speech is better with compression when SNHL is present

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

Should compression be used for severe loss?

A

Compression aids to accommodate for reduced dynamic range, (shorter release times may provide better audibility of speech)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Should compression be used for NIHL?

A

For patients with NIHL, some frequencies may not need any compression, but may need compression where the SNHL is existing (sloping moderate/severe)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

How is compression controlled or set?

A
  • We can adjust the amount of gain at soft, moderate, loud inputs and MPO in the software and manufacturer’s can have different ways of making this adjustment (we verify the gain through REMs)
  • Generally, we are not specifically selecting a compression kneepoint or choosing a ratio (although some manufacturers allow this), but changing the gain at different inputs effects the above mentioned parameters effectively alters the compression ratios and compression thresholds
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

____ settings can effect compression parameters too

A

MPO

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Raising vs. reducing the MPO

A
  • Raising the MPO – more potential room for amplitude changes
  • Reduce the MPO – tell the HA that it is going to compress to just above that gap
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What happens if a CT is set to low?

A

Too low a CT (40 dB or less) may be rejected by users because low level noise is amplified too much, but low CT may be acceptable if HA also has expansion

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

If a HA has ____ you can sometimes get away with a lower CT

A

Expansion

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

How does compression work in multichannel HAs?

A
  • Input signal is divided into multiple frequency bands/channels
  • Each channel has its own compression system
  • The CR or CT can be different within each channel to accommodate a person’s dynamic range that changes with frequency (e.g., sloping loss)
26
Q

What are we concerned about in these bands for communication and speech?

A

Speech falls across a wide range and different intensities

27
Q

Compression ratios are always in reference to ____

A

1

28
Q

Does a premium or entry-level HA have more channels?

A

Will have more channels to adjust on premium devices

29
Q

How long does syllabic compression take?

A

RT (release time) about the length of a syllable (50 to 150 ms)

30
Q

How long does phonemic compression take?

A

RT (release time) about the length of a phoneme (5 to 40 ms)

31
Q

What is spectral smearing?

A

Compression in multi-channel hearing aids can result in spectral smearing when each channel is working independently. When sound is divided spectrally into channels, amplification for some frequency regions may also fall into other adjacent areas

32
Q

What two things can spectral smearing cause?

A

1) distortion of formant information
2) loss of acoustic details

33
Q

What two things can cause spectral smearing?

A
  • More channels= more potential smearing
  • Higher CR = more potential smearing
34
Q

Speech has ____ range, from loudest vowels to softest unvoiced consonants.

A

~30-dB

35
Q

Loud ____ vowels may mask adjacent weak ____ consonants

A

Low Frequency, High Frequency

36
Q

____ compression helps by increasing the gain for weak phonemes

A

Fast-acting

37
Q

Define attack time (AT)

A
  • Refers to the time it takes to turn down the gain in response to and increase in the signal level. The attack times can vary (usually 1-20 ms)
  • 3 dB stabilization
38
Q

Define release time (RT)

A
  • Refers to the time it takes to turn up the gain in response to a decrease in the signal level. The release times can vary (20-30 ms up to approximately 5 seconds)
  • 4 dB stabilization
39
Q

How does AT and RT work in real time?

A

When a high input comes in, AT reduces the gain and stabilizes the sound, then after the sound RT amplifies the sound again

40
Q

Can you change the AT and RT on HAs?

A
  • Cannot change the programming these on most hearing aids
  • No agreement on optimal AT & RT
41
Q

What circumstances would a fast attack time be of benefit?

A
  • Door slamming, plate dropping, scream, dog barking, fireworks, for someone working on a construction site
  • Fast AT are generally preferable because it provides comfort for our patient
42
Q

What is short/fast RT best for?

A

Short RT may provide better speech audibility; with fast RT, gain recovers quicker for soft consonants

43
Q

What happens when RT is too fast?

A

Slight distortion and “pumping” may be audible

44
Q

What happens when RT is too slow?

A
  • Low gain for weak consonants. If the delay in the gain restored slowly, there may be audibility gaps for our patients
  • HF consonant could be missed because RT isn’t happening until that part is over
45
Q

When preservation on the variation between phonemes and syllables in speech is required for understanding, a release time will generally need to be within ____

A

50 ms

46
Q

What does a longer RT preserve?

A

Longer RT preserves a more natural amplitude contrast between vowels and consonants

47
Q

What kind of patients is longer RT best for?

A

Longer RT may be best for patients with cognitive impairments

48
Q

What are 6 drawbacks/disadvantages of fast acting compression?

A

1) Distortions of temporal and spectral envelope of speech
2) Fast RT tends to flatten the envelope:
3) Noise is more noticeable.
4) If noise is louder than speech, then potentially more gain for noise than speech, therefore S/N may be made worse.
5) Inter-aural level difference between two hearing aids may be reduced
- A sound from the left arriving at the right HA will be softer and therefore will receive more gain
- This affects localization or spatial awareness
6) The combination of fast RT and high CR can cause “pumping effects”
- Rapid gain changes may be perceived by the wearer

49
Q

Hearing aids can have ____ times dependant on the stimulus that triggered the change

A

Adaptive release

50
Q

What 5 patients are sensitive to temporal and structure cues?

A

1) Dependent on the Envelope of the Signal (because they can’t fill in the gaps due to cognition)
2) Low Cognition
3) Low Contextual Cues: when guessing or “filling in the blanks” are not possible
4) Elderly
5) Severe Hearing Loss: spectral cues from the signal itself may not be obtainable- more reliant on temporal cues within the signal to aid in understanding

51
Q

What are 4 advantages of slow acting compression?

A

1) Maintains the intensity relationships between speech sounds.
- Less distortions of temporal and spectral envelope of speech
- Better sound quality
2) Better speech perception (Jenstad and Souza, 2005)
3) Adults with hearing loss prefer longer RT (find sound quality more natural) (Hanson, 2002)
4) Adults with poor working memory and cognitive decline perform better with slower RT (Cox & Xu, 2014; Souza & Sirow, 2014).
- Particularly in difficult listening situations or with less redundant speech materials

52
Q

Working memory and poor memory

A
53
Q

What are 2 advantages of lower CRs?

A

1) Better speech perception with lower CR
2) Adults with hearing loss prefer lower CR
- Keep CR as low as we can when keeping the gain curves and comfort in mind

54
Q

How does adaptive RT work?

A
  • Release time depends on duration of loud signal
    • Very short (~20 ms) for sudden short intense sounds, e.g., door slam.
    • Gain rapidly returns when loud sound is over
    • Longer (~200 ms) for longer lasting loud sounds
  • Helps with comfort of loud transient sounds
  • Helps with audibility of soft sounds after a loud sound
55
Q

How does multiple systems working together work with adaptive compression?

A
  • Fast acting compression is applied to softer sounds, to help with audibility.
  • Slow acting compression is applied to medium and loud sounds, where audibility is better, to help preserve the waveform.
56
Q

What are 5 trouble shooting things to look at when a patient complains of sound distortion (“my voice doesn’t sound natural”)?

A

1) Do other people’s voice also sound distorted?
2) Do high frequencies need a boost for more clarity?
3) Is it occlusion?
4) Are CR too high?
5) Is MPO too low (insufficient headroom)?

57
Q

Based on REM/2-cc measures, what is the likely cause for this patient’s complaint (“my voice doesn’t sound natural”)?

A
  • Not much headroom
  • Try to raise the MPO in the LFs
  • If you make these changes and the patient is still complaining look at venting and see if you can adjust OE
58
Q

What are 6 things to look at when interpreting REMs?

A

1) Is REAR85 below UCLs?
2) Is average speech (65 dB SPL) meeting targets?
3) Is soft speech (55 dB SPL) audible?
4) Is loud speech (75 dB SPL) below REAR85?
5) Is there headroom between REAR with loud speech and REAR85?
6) If not, modify the CR and/or MPO.

59
Q

What are the 5 benefits of compression?

A

1) Reduces discomfort from loud sounds (through output limiting)
2) Reduces distortion of loud sounds (compared to peak clipping)
3) Increases audibility of soft consonants (if fast RT)
4) Reduces annoyance of loud transient sounds (through fast AT)
5) Reduces need for volume control

60
Q

What are the 3 limitations of compression?

A
  • Alters the temporal and spectral envelope of speech
  • Fast RT means rapid gain increases during pauses in speech, i.e., more gain for low level noise than for speech (S/N made worse)
  • May reduce ability to localize sounds
61
Q

What are 4 tips for success when setting CRs?

A

1) To help preserve speech intelligibility, don’t set the CR higher than needed!
2) Try to keep CR below 2.0:1 (this is a natural sound quality for most patients)
3) Avoid CR > 3 in hearing aids with fast release times
4) Make adjustments based on REMs and patient complaints

62
Q
A

Amplification II (13 decks)

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