Sensory Evaluation of Loudspeaker Considering Multiple Factors

Abstract

At present, many products are demanded both facilities and comforts. It is necessary for realizing comforts to measure human sensitivities. Therefore, many various sensory evaluation methods are proposed. Scheffe's paired comparison is one of these sensory evaluation methods, and is used widely. Especially, Nakaya's method, that is developed based on Schffe's paired comparison, evaluates human sensitivities and enables to add a relative score to the samples. However, the conventional Nakaya's method can separate the main effects, differences among individuals and combination effects. If there will be other factors that have an effect on the samples' evaluation, the effects will be neglected in Nakaya's method. Therefore the main effect tends to be hard to be detected by Nakaya's method. In this paper, we propose a new sensory evaluation method that enables to separate multiple effects. As this new method enables separate many effects, the relative error variance value is smaller, and it will make us to search factors easily and correctly that affect human sensitivities. This new method is applied to the loudspeakers' psycho-acoustic evaluation, and it is clarified this new method is useful for sensory evaluation when the difference is small and when there are a few panelists.