Audiologists have access to an ever-increasing number of different speech-in-noise tests for diagnostic, functional evaluation and/or screening purposes. The speech material varies widely, from digits to non-sense sentences, as does the type of masking noise (from continuous and fluctuating energetic maskers to informational maskers). When it comes to choosing the right test, clinicians need to ensure that results are predictive of the hearing ability relevant to the intended application. The current project aims to compare the performance of individuals presenting various degrees of hearing loss on three Canadian French speech-in-noise tests: a digit test (Digit Triplet Test or CDTT), a word test (Test de Mots dans le Bruit or TMB) and a sentence test (The Hearing in Noise Test or HINT). The goal was to determine (1) the spread of results for each test (e.g., range of performance, test-retest), (2) the relationship of each test to hearing thresholds, and (3) whether good performance on a particular test is indicative of good performance on the other tests. The TMB, a words-in-noise test using babble noise, yielded a wider range of results across individuals but a higher test-retest variance than the CDTT and HINT tests, which use a continuous speech-spectrum noise. Scores for the CDTT were more closely related to hearing thresholds than those of the TMB and HINT tests. Important differences in results were found across the three tests for certain individuals, indicating that selection of the speech material and/or noise type can have a significant impact on outcomes.
