Speech audiometry Flashcards
Equipment, procedures, interpretation and recording of results, speech in noise testing and its interpretation, applications for rehabilitation
Why is speech testing useful?
-Speech testing is more representative of the everyday listening experience than pure tones are
-It tests the patient’s functional hearing ability
-Predicts success with hearing aids
What equipment is needed for speech testing?
-Two channel audiometer
-Microphone and talk-back facility so that we can hear what the patient is saying
-Headphones, inserts or soundfield (using loudspeaker)
What are the speech materials that can be used for speech testing?
-Single words (can be monosyllabic or bisyllabic)
-Sentences
-Phonemes (vowel-consonant combinations)
-Nonsense words/ syllables
What is the advantage and disadvantage of using nonsense words in speech testing?
-Use of nonsense words minimises context cues and word familiarity
-However it removes the sense of language from speech testing- how effective and useful is it for real life speech?
What are the methods of presentation of the speech materials?
- Live voice
- Recorded speech
What is the advantage and disadvantage of using live voice as a method of presentation?
It is more reflective of a real world situation but you have less control over it
What are the advantages of using recorded speech as a method of presentation?
Use of recorded speech gives you more control over the stimulus and it can be standardised which increases precision and stability
Name the three types of speech testing
- Speech detection threshold (SDT)- assesses if an individual perceives speech to be present
- Speech recognition threshold (SRT)- estimates the level at which the listener is able to repeat back a word or phrase
- Speech in noise tests- provide an impression of the patient’s functioning in everyday communication situations which involve some degree of background noise
What does the speech detection threshold tell us and not tell us?
-Tells us the lowest level at which speech can be detected and identified as speech
-Does not imply that speech was understood, just that its presence is detected
When might speech detection threshold (SDT) testing be used?
-Patients who are too young to understand or repeat words
-Patients who speak another language
-Patients who have impaired language function or a speech disorder
What is the test procedure for speech detection threshold (SDT)?
-Present sentences monotonously and continuously
-Use the bracketing procedure (down 10 dB, up 5 dB)
-Ask patients to respond verbally or by pressing a button
How are speech detection threshold results interpreted?
-The result is likely to be closely related to PTA threshold for lower frequencies
-However it will not indicate a sloping loss in the higher frequencies
-Can be misleading regarding identifying overall degree of hearing loss
What is the speech recognition threshold (SRT)?
-The lowest level of sound at which 50% of the words can be recognised
-Patients needs to hear, recognise and repeat the word
What are the conditions that SRT test materials should fulfill?
-The words/ sentences should be phonetically balanced
-Patient should be familiar with the test words
-Carrier phrase may precede each word e.g. “Say the word” followed by the word
Name some examples of SRT test materials
- AB word lists (commonly used words)
- Manchester Junior Word Lists- picture presented and child matches the word to the picture
- Spondee lists- two syllable words with equal stress on both syllables e.g. hotdog, baseball
What is the recommended starting level of presentation for SRT testing and what frequencies should be presented?
-Recommended starting level: 30 dB SPL
-Frequencies: 500 Hz, 1000 Hz, 2000 Hz
How are SRT results scored? Use the AB word list as an example.
- The common word is presented to the patient and they are then asked to repeat it
- If the word ‘fish’ is presented and ‘fish’ is repeated back the patient gets a score of 3 because all three phonemes are correct
- If the word ‘duck’ is presented and ‘buck’ is repeated the patient gets a score of 2 because two of the phonemes are correct
- If the word ‘gap’ is presented and ‘pop’ is repeated then only the ‘p’ is correct and the patient gets a score of 1
- This is repeated for the whole list e.g. of 10 words and then at the end the scores are summed and converted to a percentage
How are speech recognition scores plotted on a graph?
-Speech recognition score plotted against the level at which speech is presented
-Get an S shape which is important for psychometric function
-This is known as a performance- intensity function graph
What is the relationship between presentation level and speech recognition score for someone with normal hearing?
-As the presentation level increases (gets louder) the percentage of correct score increases
-The highest score is recorded as the PB (phonetically balanced) max
-Over the PBmax the curve flattens as a plateau is reached- increasing the presentation level does not have a impact on the performance score
-Then we read off the 50% speech recognition score off the graph
Is speech detection or speech recognition easier for someone with a mild to severe sloping hearing loss? What might their performance intensity graph look like for SRT?
-Speech detection would be easier than speech recognition
-They likely would not reach 100% correct for PBmax but at a certain point increasing the presentation level would not change the correct score
Explain why the shape of the performance intensity function differs for each of these
- Normal hearing- 100% speech recognition level achieved at speech presentation level of about 30 dB above the pure tone average
- Conductive hearing loss- 100% correct speech recognition is achieved but only at an increase presentation level causing the curve to shift to the right
- Cochlear (lesion in the cochlea) hearing loss- shape of the curve changes since the patient is unable to reach 100% speech recognition
- Retrocochlear (lesion beyond the cochlea) problems- deterioration of speech as the presentation level is increased known as rollover
What is rollover and why does it occur?
-Rollover is when the speech recognition score starts to deteriorate after a certain presentation level
-At high volumes of speech presentation the words become distorted
-This is due to lesions of the VIIIth nerve
In which cases does the 50% SRT level and PTA average not correlate +/- 6dB?
- Severely sloping hearing loss- the correlation will not as robust
- Retrocochlear lesion
- Non-organic hearing loss (it is more difficult to exaggerate speech testing compared to PTA)
Why are speech in noise tests used?
-The results can be used to provide information for levels of hearing aid amplification maximum
-Can help indicate if someone needs signal processing to manage effects of background noise
-Provides a valid post-fitting measurement to evaluate intervention and quantify improvement
How are speech in noise tests advantageous over pure tone audiometry?
-Audiograms are poor indicators of speech recognition in noise
-PTA should therefore not be used solely for hearing aid selections and counselling
-Speech in noise assessment could also be used for suspected auditory processing disorder, cochlear implantation criteria and noise induced hearing loss
What are the two factors that may influence an individual’s ability to understand speech in noise?
- Age- as hearing thresholds worsen due to age, the ability to understand speech in background noise also declines (syllables, words, sentences)
- Cognition- deficits in peripheral and central processing factors cause problems with speech intelligibility
What is the second most stated reason for not wearing hearing aids?
Limitations in background noise- cannot be predicted from PTA but can be predicted from speech in noise tests more accurately
What are some examples of where speech in noise assessments could be undertaken prior to treatment?
- Patient presents with thresholds on a PTA falling within normal but report difficulties in background noise
- Pre-cochlear implantations to reduce ceiling effects
- With hearing aid options where a range of quality of hearing aids are available including analogue aids and those with minimal features
How can speech in noise tests be used after a hearing aid has already been fitted?
Can be used to verify and validate the effectiveness of the fitting and be used to improve performance of the hearing aid
Name some examples of speech in noise tests
- The Quick Speech in noise test (QuickSIN)
- Hearing in Noise test (HINT)
- Bamford-Kowal- Bench SIN test (SIN)
- City University of New York Sentences (CUNY)
What is the procedure for the QuickSIN test?
-18 unique, 6-sentence lists
-Sentences spoken by a female talker at a constant level in a background of 4-talker babble
-Target sentence either presented through the same loudspeaker as the babble or through two loudspeakers
-The babble increases in 5 dB steps from +25 to 0 dB
-The patient is asked to repeated the sentences presented
-Each correctly repeated word is awarded one point
-22.5- score of total words correct= SNR loss
What are the advantages and limitations of the QuickSIN test?
Advantages: Easily available, easy to set up and administer, easy calibration, quick completion, straightforward scoring method, can be adapted for assessing benefits of directional microphones
Limitations: Certain lists are only appropriate for American English, 2-3 lists need to be used to obtain an average test score
What is the signal to noise ratio required for a normally hearing person and a hearing impaired person to hear the target talker over the background babble?
Normally hearing individual: Requires +2 dB SNR (target speaker needs to be 2 dB louder than the background babble to correctly repeat 50% of the key words)
Hearing-impaired person: +12 dB corresponds to 10 dB SNR loss
What amount of SNR loss constitutes a normal hearing, mild hearing loss, moderate hearing loss and severe hearing loss (QuickSIN test)?
Normal: 0-2 dB
Mild: 3-6 dB
Moderate: 7-12 dB
Severe: >12 dB
Describe the test procedure for the hearing in noise test (HINT)
-250 Bench-Kowal- Bamford (BKB) sentences
-Sentences of 5-7 syllables presented by a male speaker
-Presented either from the front, right or left
-Competing noise held constant at 65 dB
-Ascending approach used to determine the presentation level and subsequent presentation levels are increased/ decreased in 4 dB steps and then 2 dB steps
-dB SNR= RTS (level at which listener can correctly repeat 50% of the sentences)- dB noise
What are the advantages and limitations of the HINT test?
Advantages: Can differentiate small differences amongst people and products, one of the most researched speech tests
Limitations: More complicated set up, test administration and scoring, clinics need to obtain their own sound field norms
What is the test procedure for the Bamford-Kowal-Bench SIN test?
-10 short sentences in each list with 18 equivalent pairs
-Multi-talker babble ranging from +21 dB to -6 dB in 3 dB steps for the 10 sentences
-Target sentences presented with the babble in one loudspeaker or separately from two loudspeakers
-Goal is to track improvements in SNR at which listener can achieve 50% sentence recognition
-SNR-50: 23.5- number of key words in each sentence, then average values for both lists
What are the advantages and limitations of the Bamford-Kowal-Bench SIN test?
Advantages: Quick and easy to administer and score, separate lists for cochlear implant users and those with severe hearing loss or significant SNR loss, can be used for all adults
Limitations: Less normative data
What are the SNR loss values associated with normal hearing, mild hearing loss, moderate hearing loss and severe hearing loss (Bamford-Kowal-Bench SIN test)?
Normal: 0-3 dB (might hear better in noise than those with normal hearing with hearing aids)
Mild: 3-7 dB (may hear almost as well in noise as those without hearing loss)
Moderate: 7-15 dB
Severe: >15 dB
Describe the procedure for the City University of New York Sentences (CUNY) test
-24 sentences varying from 3-14 words in 12 lists
-Visual component available
-SNR during testing is presented at +10 dB to represent an environment more typical of everyday listening conditions
-Score is given for how many words, keywords and full sentences are correctly identified
What are CUNY tests most traditionally used for?
-As an assessment tool within cochlear implant clinics to assess the ability of lip-reading
-Used to help compare pre- and post- fitting information with service users including additional rehabilitative strategies to amplification
What are the advantages and limitations of the CUNY test?
Advantages: Easy to score, has audio-visual component
Limitations: Ceiling effects more common, can be too easy, lengthy test (24 sentences per list) with several conditions
