Speech Audiometry: at Biswajeet Sarangi, B.Sc. (Audiology & Speech Language Pathology)
Speech Audiometry: at Biswajeet Sarangi, B.Sc. (Audiology & Speech Language Pathology)
SPEECH AUDIOMETRY
Pure tone Audiometry provides only a partial picture of the patient’s auditory sensitivity.
Because it doesn’t give any information about it’s ability to hear and understand speech.
To find out the person’s ability to hear and understand speech. We do this testing with
speech stimulus and this process is called Speech Audiometry.
According to ANSI S-36 1996 – The characteristics of audiometers including those used
to do speech audiometry are given in the ANSI specification for audiometry.
Here are the process which are involved in the speech audiometry or these are the
components which are required for the speech audiometry process for an individual.
These are the following :
Input Selection :
To select the desired source of speech material
Tape, deck, CD player = For recorded speech.
Microphone = For live voice.
Which is used with the V-U (Volume Unit) meter to ensure that the speech signal
are of at the level necessary for them to being properly calibrated.
Attenuation :
Output selector :
Through which mode you want to present the speech signal to the patient.
! CARRIER PHASE : It is the phase which proceeds the stimulus word, during
speech audiometry. It is designed to prepare the patient for the test word and to
access the clinicians fault, controlling the input loudness of the test word.
! SRT : (Speech Reception Threshold). The threshold of a person for speech at the
lowest level at which the presence of speech signal can be heard or recognized or
identified 50% of the time.
! SDT : (Speech Detection Threshold). The lowest level at which the presence of
speech signal can be detected 50% the time is called as SDT or speech awareness
threshold.
! SPONDIAC WORDS : It is 2 syllable word pronounciated with equal stress on
both the syllables. E.g. “dhoop-chhawn”, “khel-kud”, “daud-dhup”.
! SPONDIAC THRESHOLD :The lowest level at which 50% of the spondiac
word is correctly identified.
! UCL : (Un Comfortable Level). It is sound pressure level at which the speech
become uncomfortably loud.
# To serve as a measure for cross checking the pure tone threshold. ( SRT & PTA
co-relation is +/- 12dB)
# To determine hearing aid needs and performance.
# To evaluate the functional hearing loss/ malingerers. If the patient has Functional
hearing loss then the difference with SRT may be more or less than 12 dB. If
more than 12dB = conductive hearing loss. If it’s less than 12dB then = sensory
and If lower than that then the hearing is neural hearing loss.
$ Spondaic words : Most of the SRT are obtained using the Spondaic words.
$ Sentence Test : Although spondaic words are used usually for SRT but
they are not only the materials for SRT purpose. Sentences are the principle
alternative used for SRT testing, especially in noise. Given by PLOMP and
MIMPIN – 1979.
$ Other materials used :
# Monosyllabic words
# Disyllabic words
# Paired words
# Sentences with key words
Each sound consists of particular frequency and to perceive the particular frequency,
intensity is required. Each sound in each place is different. Noise differs in terms of
manner of articulation rather than the place of articulation. Each sound has got 3
parameters :
• Frequency
• Intensity
• Duration of sound
These 3 are different for each speech sounds. It differ from individual to individual, time
to time, and also varies depending upon proceeding sound and following sound.
Fricatives
Normal
Stops
Vowels
# Fricative = High Frequency Stop, Vowel = Mid Frequency Nasal = Low frequency
But the speech concentration energy is not same. Maximum energy concentration will be
around 2KHz. At 4KHz = 50dB, 1KHz = 52dB, and at 500Hz = 26dB. Normal
conversation falls within 50 – 60dB.
The PTA doesn’t talk about the communication ability of the person. Communication
ability of two persons are not same. Thus speech audiometry helps in different aspects
such as :
• To cross-check the pure tone threshold.
• To find out the type of hearing loss
• To find out the degree of hearing loss
• Help in hearing aid selection
• Help in identifying functional hearing loss
• Helps in identifying the site of lesion.
You can find out SDT under the head phone, or in free field condition. You will always
get response from the better ear. This is the main disadvantage of caring out SDT in free
field condition.
Input through the microphone and output through the headphone is used. Then we present
monosyllabic words like papa, mama, baba, nana, at the intensity above threshold level.
When you are presenting the stimulus make sure that the microphone is “ ON ”, then
follow “Up & Down” method like Pure Tone Audiometry. Make sure that you are getting
responses twice at a time.
SDT Vs PTA
# Spondaic words
# Paired words
# Sentences
# Conversations
# Cold Running speech ( Speech which doesn’t change intonation )
SRT Vs SDT
(Difference between SRT & SDT is about 9 dB )
# SDT is always better than SRT because in SDT they just have to identify or
detect the speech signal which is present while in SRT they have to
comprehend the speech signal & then have to repeat the speech stimulus.
SRT Vs PTA
# SRT correlates more with the Pure Tone Average rather than the individual
pure tone threshold.
# In case of rising type of audiogram take the PTA at the two best frequencies
and the SRT will correlate with the threshold of those frequency.
# In case of high frequency sloping we do the same thing.
# SRT & PTA should always correlate +/- 12dB, given by Ventry & Chaiktin –
1961. Again Olsen said that SRT & PTA correlate +/- 6dB. But for clinical
measurement we take +/- 10dB.
You can present speech stimulus through bone vibrator using the speech audiometer
using the same speech materials.
ASCENDING METHOD :
e.g.
PTA = 50dB ( Present intensity for below PTA )
After matching the trials, which 2 trials have the same intensity, that will be SRT of the
person. Hence in this case SRT is 45dB
Procedure :
# Present the stimulus below the threshold level and increase the intensity and find
the threshold.
# Set the intensity below the PTA threshold.
# Present one spondee – No Response
# Then increase the intensity by 10dB and present the stimulus (One spondee). You
may get a response or may not get.
# If there is no response then increase the intensity by 10dB and present the
stimulus. Keep increasing the intensity by 10dB until you get a response. If you
get a response at a particular intensity, at that intensity you decrease the intensity
by 15dB and that reduced intensity will be Initiation level.
# Here you are presenting 4 spondees. You will not get any response at this level.
# Now increase the intensity by 5dB and see whether you got response or not. You
have to increase the intensity by 5dB and present 4 spondees till you get the
response of 2 out of 4 spondee. That is the end of the 1st Trial.
# Then reduce the intensity by 10dB and this level will be initiation level for the 2nd
ascending trial. Present 4 spondee. You may or may not get any response.
Increase the intensity by 5dB and present 4 spondee. Check you get response or
not.
# You have to increase the intensity by 5dB and 4 spondee till you get any response
for 50% times and that’s is the 2nd Trial.
# Now reduce the intensity by 10dB, same process you put.
In all the three trials you check which 2 trials are same. The 2 trials which are same that
will be the SRT. If all the 3 intensities are different then go for the 4th trial.
DESCENDING METHOD :
e.g.
50dB – 2 spondee – R
decrease by 10dB SRT = IL – CR + ½ step
40dB – 2 spondee – R
decrease by 10dB SRT = 30-5+1
30dB – 2 spondee – R = 26dB
decrease by 10dB
20dB – 2 spondee – NR IL = Initiation Level
20dB – 2 spondee – NR CR = Correct Responses(+ve)
increase by 10dB
30dB – 2 spondee – R – initiation Level
decrease by 2dB
28dB – 2 spondee – R
decrease by 2dB
26dB – 2 spondee – NR
Procedure :
# Present the stimulus or intensity for above the pure tone average and try to get
50% response. Here you are presenting 2 spondee.
# Take the intensity and present the stimulus. Here you will surely get the response.
If you are getting the response then decrease by 10dB.
# Again you present the 2 spondee, you may or may not get a response.
# If you get the response then decrease by 10dB. If you are not getting the response
then it is negative.
# At this level again you have to present 2 spondee to cross check., then increase by
10dB and you will get a response then this is called initiation level.
# From here decrease by 2dB and present the same 2 spondee and check whether
you get a response or not.? Keep on decreasing by 2dB until you find there is a
No Response. Stop where you don’t get any response.
Present the stimulus above the pure tone average. Probably at PTA+20dB. Present 3
spondee words. See for at least 50% responses.
SRT is that in which at least 2 intensities matches equally.
e.g.
PTA = 40dB
Start with PTA+20dB = 60dB
60dB – 3 spondee – R
1st Trial = 35dB
decrease by 10dB
2nd Trial = 30dB
50dB – 3 spondee – R
3rd Trial = 30dB
decrease by 10dB
40dB – 3 spondee – R
2nd & 3rd Trials are equal
decrease by 10dB
Hence the SRT here is
30dB – 3 spondee – NR
30dB.
increase by 5dB
35dB – 3 spondee – R -------------------------- ( 1 )
decrease by 10dB
25dB – 3 spondee – NR
increase by 5dB
30dB – 3 spondee – R -------------------------- ( 2 )
decrease by 10dB
20dB – 3 spondee – NR
increase by 5dB
25dB – 3 spondee – NR
increase by 5dB
30dB – 3 spondee – R -------------------------- ( 3 )
Procedure :
Initially you present the stimulus above the pure tone average. Probably at PTA+20dB.
At this level you should present 3 spondee words. Present 3 spondee at this level and see
for the response. Some people may say that present the stimulus and see for the
response. i.e. if you get response then reduce the intensity by 10dB and see for response
again. If you are not getting the response by at least 50% then increase the intensity by
5dB and present 3 spondee. If you are getting the response the make the note of the
intensity i.e. the Initiation Level. Then decrease the intensity by 10dB and present the
stimulus. If you are not getting the response the again increase by 5dB and present the
stimulus (3 spondee ). If you are getting response at this level then that will be the 2nd
ascending trial. Then decrease the intensity by 10dB and present 3 spondee. If you are
not getting the response then increase the intensity by 5dB and see for the response. If
you are getting the response at this level then this will be the end of 3rd ascending trial.
And check for all the 3 trials and see which two values are same and they will be the
threshold.
This is to find out speech discrimination ability of an individual. This score is doene at
supra threshold level. This is calculated in terms of percentage.
i.e.
Number of responses
----------------------------------------------------- X 100
Total Number of speech stimulus presented
This should come 27 – 30dB above SRT in normal then it reaches a platue. If you will
increase intensity more than this you will not get any variation.
Materials Used :
% PB words ( Phonetically Balanced Words )
% Simple words
% Simple sentences with key words
In normal = 100% normal response, 27 – 30dB above the SRT score.
Conductive hearing loss cases will have better graph than SN hearing loss cases. Because
SN people have discrimination problem.
Normal 100%
SN hg.loss
70% – 80%
As you keep on increasing the speech stimulus, the lower the score will be due to neural
fatigue.
In SN hearing loss,
Better the hearing loss, more is the performance.
More is the hearing loss lesser is the performance.
Sensory = Cochlear Pathology, Neural = Retro-Cochlear Pathology.
Jerger & Jerger (1977) gave a speech test to differentiate between CP & RCP, i.e. PI–PB
function ( Picture Identification – Phonetic Balance )
A different SL level i.e. SRT+SL. You have to find out the speech identification score.
You need to use PB list and find out maximum & minimum score. If the difference is
more than 0.45 then it is RCP and if the difference is less than 0.45 then it is CP.
! Familiarity of the word ( Educated persons will know the words and they will
respond easily & quickly. If uneducated, then they may fail to respond. So
educated people will have better performance than others )
! Practice Effect ( If a person comes regularly for the assessment then he may get
familiar, which may show better response and that will affect SRT )
! Live Vs Recorded version ( Live when the clinician uses his own voice. Recorded
when clinician records his voice and present the stimulus through tape recorder.
In recorded voice the player produces some kind of mechanical noise. So, it’s
always better to go for live voice rather than recorded.
! Visual clues
! Close Vs Open set ( closed set is using head phone, Open set is free field
situation. In open set patient respond through better ear. So closed set has better
response than open set.)
! Carrier Phase ( means the instruction or phase which proceeds the stimulus words
during speech audiometry. Designed to prepare the patient for the test)
! Types of hearing loss ( Conductive performs better than Sensory. Sensory
performs better than Neural)