Purpose: Auditory training using multiple-choice items allows easy manipulation of difficulty level, efficient scoring, and analysis of incorrect response patterns. The purpose of this study was to develop multiple-choice sentences as a closed-set material for auditory training. Methods: Each sentence was a five-word sentence (name, adjective, object, numeral, verb). Each of the 5 words contained 10 alternatives, resulting in 50 response alternatives (10 names, 10 adjectives, 10 objects, 10 numerals, 10 verbs). The construction of sentences, the selection of target words and speaker as well as the processes of recording, cutting and resynthesis were based on the recommendations of the International Collegium of Rehabilitative Audiology. In total, 1,500 sentences were generated from a combination of the 50 response alternatives. Results: Twenty-two adults with normal hearing participated in the preliminary evaluation of multiple-choice sentences. After hearing each five-word sentence, listeners were required to choose the five words they had heard from the 50 response alternatives. To verify the applicability of those closed-set sentences in the background noise, evaluation was conducted in quiet as well as under background noise conditions. Results showed that the multiple-choice sentences developed in this study were identified with mean accuracy greater than 95% in all conditions. Conclusion: Our ten-alternative multiple-choice sentences could be used as stimuli for closed-set auditory training. The multiple-choice sentences could be encouraging especially for hearing-impaired listeners who have difficulty in receiving open-set auditory training. Further research is needed to determine the clinical feasibility of our multiple-choice auditory training for older people with hearing loss.