Methods Final

basic levels of speech processing

detection, discrimination, recognition, comprehension

long term speech spectrum

visual representation of where the sounds of speech are in terms of pitch and volume ; most intense around 500 Hz, range of intensities = 30 dB

*shaped from the ISTS (international speech test signal)

What is the ISTS

international speech test signal

This is bits of speech from six females speaking different languages that are blended together to form unintelligible speech

what is the long term speech spectrum referred as to patients?

the speech banana

articulation index/speech intelligibility index

percentage of the sounds within speech that someone can hear ; proportion of LTSS that is audible, ranges from 0-1.0, 1= 100% of LTSS, 10-50 dB

everyday sounds audiogram

shows which sounds are associated with a certain pitch and loudness on an audiogram, sounds below dashed line are audible and sounds above dashed line are inaudible

mlv

monitored live voice, significant inter and intra-tester variability, mlv score was not within the range of variability for recorded speech scores

what does audiometric zero for speech represent?

the average speech threshold for NH listeners in dB SPL when the dial is set to 0 dB HL

speech detection threshold

Lowest dB HL at which pt. signals presence of speech at least 50% of the time

speech reception threshold

Lowest dB HL at which pt. correctly identifies 50% of a set of spondaic words

what is the name of the spondee list?

CID- W1

Modified Chaiklin and Ventry SRT

familiarize with 15-20 spondees

1. Present 1st at 25 dB above FA & up to 6 spondees to determine if hear 3/6

2. If so descend in 5 dB steps with up to 6 at each level

3. Once below 50% stop & record lowest level at least at 50%, as SRT

Martin and Dowdy SRT

familiarize with 10-12 spondees

ASHA procedure with spondees rather than tones

ASHA 1988 SRT

1. Start 30/40 dB above PTA;

2. Present 1 spondee & if correct drop 10

3. When word missed present 2nd word at that level

4. Descend in 10 dB steps until level where both words missed

5. Go up 10 dB from stop level & present 2 spondees.

6. Go down in 2 dB steps until last 5/6 words are incorrect.

7. Calculate threshold.

SRT = SL (starting level) - # correct +1

Brandy PTA-SRT agreement values

Within 6 dB is "good agreement"

7-12 is fair agreement

12+ is poor agreement

types of speech stimuli

nonsense syllables, monosyllabic words, sentences, spoken paragraph

open set vs closed set

open = client is not given set of possible syllables, words, or sentences

closed = client is given the set

options for presentation level

1. Average level for conversational speech

2. MCL for client

3. Intensity level to estimate maximum performance

4. Use multiple presentation levels

Makersie et al Presentation levels

Set the WRS presentation level 5 dB below speech UCL, or

Set the attenuator dial above the 2 kHz threshold based on:

If 2 kHZ threshold < 50dB HL, add 25 dB SL

If 2 kHz threshold = 60-65 dB HL, add 15 dB SL

If 2 kHz threshold = 70-75 dB HL, add 10 dB SL

NU-6 Word Lists

Lehiste and Peterson (1959) developed a set of 500 CNC words

Tillman & Carhart took this word set and developed 4 lists

Known as NU-6 lists—widely used today

Maryland CNC word list

Took 500 of the original CNC words and embedded them in the carrier phrase "Say the ______ again." (schwas on either side of the stimulus words to control coarticulation effects).

10 lists of 50 words

QuickSin

Each test list contains 6 IEEE sentences

Syntactic cues but subtle semantic cues

Multispeaker babble with SNRs from +25 to 0

SNR loss

the dB increase in signal-to-noise ratio required by a hearing-impaired person to understand speech in noise, compared to someone with normal hearing

performance intensity function (PI)

shows how the patient's speech recognition performance depends on the intensity of the test materials

rollover index

a reduction of speech recognition scores that occur at intensities above the level where PBmax is obtained (associated with retrocochlear disorders)

PB max-PB min/PB max

Rollover index of .25 or greater indicate a retrocochlear problem (for NU-6 test)

thorton and raffin 1978

Generated 95% CI for WRS to demonstrate normal variability

Shorter word lists result in more variable scores

Can use arrows on SPRINT chart for T & R, arrows represent normal variability

Dubno et al 1995

data represented by shading on spring chart, offered 95% confidence limits for maximum SR scores for clients with NH or SNHL of varying degrees of HL

masking rule for SRT

SRT level in TE - IA > BBC of NTE

IEM for SRT masking

SRT in NTE + 10 dB = IEM

4 words at each level

Masking rule for supra-threshold word recognition

PL in TE - IA > BBC of NTE

how to determine the masking level for supra-threshold

PL - IA + largest ABG in NTE + 10 dB SF

masking for air conduction rule

AOBG > IA

IEML air conduction

AC NTE + 10 SF

masking for bone conduction rule

ABG > 15 dB

IEML bone conduction

AC NTE + OE + 10 SF

no plateau check for "O"

dB EM NTE - IA > BC TE

dB SL

sensation level, the intensity of sound above above an individual's own threshold

RETSPLs

Setting for 0 dB HL on the audiometer produces the SPL at that frequency that corresponds to average NH at that frequency for that specific transducer

- These SPL are called RETSPLs.

dB HL

used for diagnostic testing

dB SPL

used for hearing aid fitting

biological listening check

to test attenuator dial, set frequency at 1000 Hz

to test frequency dial, set attenuator at 70 dB HL

checking for cross talk

Direct a 70 dB HL 1000 Hz tone to right earphone and disconnect that earphone listening for cross talk in the left earphone; direct the tone to the left earphone and repeat the procedure; tone should not be heard in the opposite earphone

AAOO percent hearing impairment formula

Calculate AAOO PTA (.5, 1, 2, & 3);

PTA - 25 dB (low fence) = a;

A x 1.5% = percent HH for that ear;

goodman degrees of hearing loss

0-25 = normal

26-40 = mild

41-55 = moderate

56-70 = moderately severe

71 - 90 = severe

91+ = profound