[200 years testing hearing disorders with speech, 50 years german speech audiometry -- a review]

FROM THE BEGINNIGS TO CLASSICAL SPEECH TESTS: The need for classification of different degrees of hearing disorders first arose when it seemed possible to treat deafness. Grapengiesser in Berlin 1801 had applied galvanic current to the ears of deaf children and reported some success. Pfingsten in Kiel in 1804 using this method was the first to use speech in diagnosing different degrees of deafness. He divided speech sounds into three classes: vowels, voiced consonants and voiceless consonants. Itard in Paris in 1821 gave a classification of five classes according to which sounds could be perceived, starting from normal speech to thunder and the bang of a gun. Schmalz in Dresden 1846 noted the range within which speech was understood thus introducing the concept of hearing distance. Helmholtz in 1863 had demonstrated that vowels are composed of pure tones. Wolf in Frankfurt 1871 tried to align all speech sounds from the lowest frequency (tongue-R = 16 Hz) to the highest (sh = 4096 Hz) and measured the hearing distance for each sound. Following these suggestions word lists based on the predominant frequencies were compiled in a number of languages including Japanese.

This chapter is devoted to Karl Heinz HahIbrock, Freiburg, who was the founder of the German speech audiometry. Hahlbrock followed the American authors of the Psycho-acoustic Laboratory at Harvard, in particular J. P. Egan (1948), using statistical methods for composing lists of words based on the relative frequency of speech sounds and phonetically balanced between the different groups. He finally presented a test comprising groups of two-digit numbers and monosyllabic words. Hahlbrock died in 2003 exactly fifty years after the presentation of his test. A short account of his life is given.

In the following years various other types of speech tests were elaborated using sentences, distorted speech, diotic and dichotic presentation partly aimed at fitting hearing aids, partly with the aim to diagnose central hearing disorders. Hahlbrock's test, however, remained the standard for evaluating speech reception and discrimination.

One of the fundamental problems in testing speech discrimination is that there is no catalogue of phonemes common to all languages or regional accents. Untrained not native speakers often do no perceive certain sounds having a partial auditory agnosia. They cannot distinguish between e. g. 'hand' and 'and' or 'end' and 'ant', but the examiner must decide if the word presented was repeated correctly or not.