The purpose of this study is to develop compensation methods for the hearing disorders of hearing-impaired persons listening to music as well as to speech. In this study the difference in effectiveness of compensation methods between speech and music was investigated. Several compensation processes (equalization and compression) were subjectively evaluated with adjective pairs in both speech and music. In this experiment the signal processing was run only for simulating a certain hearing disorder including recruitment characteristics. The results show that ``naturalness'' and ``easiness'' relate to overall evaluation in music perception, while ``clearness'' relates to overall evaluation in speech perception.