| Summary: | 碩士 === 大同大學 === 工業設計學系(所) === 103 === Human language is an important way to communication. In addition to passing by text message, the voice also can strengthen the emotional communication. With advances in technology many products also joined the voice prompt function, this is for the visually impaired and reading disabled persons helpful, It also simplifies the inconvenience and ancillary products operating on visual operation of the product. This study evaluated the experiment by the imagery of semantic differential analysis, to understand the different properties of subjects who, for speech imagery evaluate differences, and to explore the various features of relevance between the voice and image of perception.In the early experiments, we did survey on 63 subjects, of whom 29 men and 34 women, according to the literature by the subjects were selected from 15 groups for 32 groups of adjectives. Choose a suitable degree of synthetic speech Likert scale.The spindle of latter experiment we use 15 sets of adjectives for nine speech samples on semantic differential analysis, subjects were 35 people in a One-sample ttest for gender comparison, then ANOVA test differences in age, occupation and education, and finally the voice samples principal component analysis, Wherein, when the pitch decreases, it will give people more kindly feelings, but at the same time will make people feel more low-key voice, weak and seems to be rather sad, in the voice performance is easy to be turbid, more visibility for voice poor. Although the speed of sound under the same state, deep voice will have to speak very slowly feeling. Students and freelancers for lower sound feel more strongly than the deep voice students think there is elegant, with a nice feeling of happiness, freelancers are feeling is exactly the opposite.
|
|---|
