The ultimate goal of this study is to develop a violin-playing robot that can produce expressive musical sounds for various impressions for human listeners to enhance human-robot communication. In this paper, we propose design principles of loudness for an entire musical piece, phrases, and one note in short musical pieces to express bright and dark timbres. Bright and dark timbres are opposite to each other and considered to be fundamental musical expressions. We conducted an experiment in which a human violinist played short, famous, and easy musical pieces and the produced sounds were recorded and analyzed by the FFT method. Also, we interviewed the violinist and listened to the sounds to identify the design principles of loudness. Based on our experimental results, interviews, and listening, we proposed several design principles on loudness in a musical piece. For example, inserting an interval between phrases gives a bright impression. On the other hand, the slope of the loudness at the beginning of musical pieces in a dark timbre should be smaller than in a bright timbre.

This paper describes a method for finding structural matching between parallel sentences of two languages, (such as Japanese and English). Parallel sentences are analyzed based on unification grammars, and structural matching is performed by making use of a similarity measure of word pairs in the two languages. Syntactic ambiguities are resolved simultaneously in the matching process. The results serve as a useful source for extracting linguistic and lexical knowledge.