How Speech Reception Threshold (SRT) is Measured
Speech reception threshold (SRT) is measured as the minimum hearing level (signal-to-noise ratio) at which an individual can correctly recognize 50% of presented speech material, typically using an adaptive procedure that adjusts stimulus levels based on the patient's responses to successive sentences or words. 1, 2
Core Measurement Methodology
The SRT test is conducted independently for each ear using the following approach:
- Starting point: The initial sound level is determined from pure tone thresholds (PTTs) obtained via pure-tone audiometry, typically using the pure tone average (PTAv) 1
- Adaptive procedure: The signal-to-noise ratio (SNR) of successive sentences is adjusted up or down based on the subject's scores on previous sentences 2
- Endpoint determination: The SRT is estimated as either the mean of a subset of the SNR levels tested, or by fitting a psychometric function to the data 2
Clinical Application Context
In clinical practice, particularly for sudden sensorineural hearing loss assessment:
- Recovery assessment: Complete recovery is defined as follow-up SRT improving to within 10 dB of pre-hearing loss levels 3
- Partial recovery: Follow-up SRT improves to within 50% of pre-hearing loss levels 3
- Significance threshold: Any change must exceed 10 dB to be considered clinically significant, as smaller changes fall within test-retest reliability limits 3
Key Measurement Parameters
Fixed vs. adaptive noise levels: SRT can be measured at either a fixed absolute noise level (e.g., 80 dBA) or at a fixed sensation level (e.g., 25 dB above threshold), with the choice affecting comparability between normal-hearing and hearing-impaired listeners 4
Test reliability factors:
- Patients with poor speech discrimination yield worse SRT reliability and greater sensitivity to the starting hearing level 5
- The number of repetitive adjustments directly impacts both patient and audiologist fatigue, potentially compromising test reliability 1
Optimization Strategies
To improve efficiency and reduce testing time:
- Initial level estimation: More accurate initial sound level estimation (closer to the final SRT value) reduces the number of repetitions required 1
- Conversion methods: When tests use steady-state masking noise matching the speech spectrum, SRT values can be converted to percentage-correct scores and vice versa without requiring pure tone threshold information 6