Speech Recognition Threshold
Purpose
The Speech Recognition Threshold (SRT) is the lowest intensity level at which a patient demonstrates the ability to identify speech 50% of the time. Identification typically is assessed by repetition of words. This can be done through other methods such as pointing to pictures.
The words used for determining SRTs are known as spondees. Spondees are two-syllable words that have equal stress on each syllable. Examples are words such as baseball, cowboy, hot dog, sidewalk, airplane, playground, ice cream, popcorn, football and cupcake.
The use of monitored live voice (MLV) versus use of recorded materials has been found to have little effect on the outcomes of speech threshold testing. Because of this determination, and because the use of MLV is more efficient for establishing SRT, MLV is typically the chosen method. However, in cases where the tester has speech that may be perceived as heavily accented relative to that of the tested population, recorded materials should be used.
The words used for determining SRTs are known as spondees. Spondees are two-syllable words that have equal stress on each syllable. Examples are words such as baseball, cowboy, hot dog, sidewalk, airplane, playground, ice cream, popcorn, football and cupcake.
The use of monitored live voice (MLV) versus use of recorded materials has been found to have little effect on the outcomes of speech threshold testing. Because of this determination, and because the use of MLV is more efficient for establishing SRT, MLV is typically the chosen method. However, in cases where the tester has speech that may be perceived as heavily accented relative to that of the tested population, recorded materials should be used.
Materials
Audiometer with the use of monitored live voice (MLV) via a microphone, transducer and a spondee list.
The presenters voice should be modulated to peak at zero on the VU meter.
15 Most Homogenous Words In Audibility
|
|
|
19 Most Easy To Picture
|
|
|
|
Procedures/Instruction
To use monitored live voice, the tester must present words at a consistent intensity level. The tester must monitor the input by examination of the VU meter on the audiometer to ensure that speech stimuli are neither too loud nor too soft. It is common practice to utilize a carrier phrase such as "Say the word..." prior to presentation of the spondee. The presenters voice should be modulated to peak at zero on the VU meter.
Because patients may perform more poorly if they are unfamiliar with a particular word, it is good practice to
The threshold level can be determined by
Because patients may perform more poorly if they are unfamiliar with a particular word, it is good practice to
- familiarize the patient with the word list prior to initiating testing. This can be accomplished by finding a comfortable listening level and presenting the words to the patient at this level.
- Once the patient has had the opportunity to repeat the words at a comfortable level, the intensity can be reduced for the purpose of determining the threshold level.
- It is crucial that the patient not see the face of the audiologist during this testing, if it is conducted using monitored live voice. Therefore, the audiologist must either be out of view of the patient, or must completely cover her or his face to avoid presentation of any visual speech cues
The threshold level can be determined by
- presenting the words at a comfortable level to familiarize the patient with the word list
- The level is then decreased to 30 dB HL and a single word is presented
- If the patient responds correctly at 30 dB HL, the intensity is decreased by 10 dB.
- This process is continued with presentation of a single word until the patient fails to respond or responds incorrectly
- If the patient does not respond or responds incorrectly at 30 dB HL, the intensity is increased by 20 dB
- This process is continued with presentation of a single word until the patient responds correctly.
Interpretation/Site of Lesion
Because speech is used to determine speech thresholds, the thresholds of the pure-tone frequencies that are most important for speech understanding should closely match the SRT. Specifically, an average of the pure-tone thresholds at 500, 1000, and 2000 Hz (known as the pure-tone average or PTA) should match the SRT within 7-10dB. When SRT measures do not closely match the PTA (with the SRT measure tending to be lower in intensity than the PTA) this may indicate functional or exaggerated hearing loss.
There are exceptions to the rule that the SRT and PTA should closely match, such as in cases of highly configured losses. For example, if 500 and 1000 Hz thresholds are 20 dB HL, but hearing at 2000 Hz drops to 80 dB HL, rather tan the PTA of 40 dB HL.
Regarding the speech detection threshold, the intensity level of the SDT should correspond to the lowest intensity level of the pure-tone frequencies tested, rather than the PTA.
There are exceptions to the rule that the SRT and PTA should closely match, such as in cases of highly configured losses. For example, if 500 and 1000 Hz thresholds are 20 dB HL, but hearing at 2000 Hz drops to 80 dB HL, rather tan the PTA of 40 dB HL.
Regarding the speech detection threshold, the intensity level of the SDT should correspond to the lowest intensity level of the pure-tone frequencies tested, rather than the PTA.
Management
Works Cited
Gelfand, Stanley A. Essentials of Audiology. Thieme, 2016.
DeRuiter, Mark, and Virginia Ramachandran. Basic Audiometry Learning Manual. Plural Publishing Inc., 2017
Image of sounds of speech: www.michaelidesvermaas.co.za/sites/default/files/pictures/sounds-of-speech.jpg
DeRuiter, Mark, and Virginia Ramachandran. Basic Audiometry Learning Manual. Plural Publishing Inc., 2017
Image of sounds of speech: www.michaelidesvermaas.co.za/sites/default/files/pictures/sounds-of-speech.jpg