SUMMARYIn this paper, we study improving the noise robustness of formant frequency extraction based on linear predictive analysis by using the noise reduction ability of the autocorrelation function. The autocorrelation function has the property of concentrating the noise components in the short delayed parts of speech signals corrupted by white noise. We believe that the noise robustness of formant frequency extraction can be improved by utilizing the above property and using the autocorrelation function of the speech signal instead of the signal itself in formant frequency extraction. In this paper, we analyze the principle behind extracting the formant frequency from the autocorrelation function and examine this problem and a possible solution in detail. Then based on this analysis, we propose Method 1 that performs linear predictive analysis of the autocorrelation function of the speech signal. Then we verify from the experimental results that an extraction accuracy at the same level as the conventional method is obtained for a clean signal by Method 1, and the extraction accuracy is significantly improved over the conventional method for speech corrupted by noise. However, inadequate extraction accuracy is also indicated in a very noisy environment, and we analyze the cause and propose Method 2 as an improved method that subtracts the autocorrelation function. The experimental results show that when the signal-to-noise ratio is 15 dB or less, the formant frequency extraction error (Average Absolute Error) in Method 2 is kept to about one-third the error of the conventional method and about one-half the error of Method 1.

No abstract

This paper describes hybrid relations in relational databases that allow existing relation to be altered by the addition of new attributes without reorganization of the database schema. The values of new attributes with respect to an existing relation are stored separately from the relation as a set of triples of tuple identifier, attribute name, and value. At query time, a hybrid relation, which has only the attributes requested in a query, is derived virtually by combining the relation and this set of triples. A relation can be reorganized by upgrading its attribute values from these triples. The hybrid relation is defined as an algebraic expression, and equivalent expressions of a query on the hybrid relations are shown for efficient query processing.

The long-term generation scheduling in power utilities is aimed at maintaining power supply sufficiency and estimating fuel consumption. Multiperiod constraints, such as the allowable number of unit commitments of thermal units and fuel consumption with a given total amount of fuel, must be taken into appropriate consideration for practical economic scheduling in the long-term scheduling horizon. Due to the large size and complexity of such scheduling problems, it is difficult to obtain solutions collectively in terms of stability and processing time in operational situations. We propose a new scheduling method which consists of "preparative processing" and "detailed optimization processing." The former acts to divide the longterm problem with multiperiod constraints into time units of a week according to quick and simplified scheduling results, and the latter acts to minimize the total operation cost on the basis of those weekly partitioned problems. The processing starts with an optimal scheduling result excluding the aforementioned two multi-period constraints and proceeds to sequentially resolve the violation of each constraint with quantitative consideration of interrelationships between them. This paper describes the validation and effectiveness of the proposed method through the real world example of Chubu

Multimedia information access on the Internet creates a new paradigm for museum information and education service that complements conventional school programs.We designed and developed the Global Digital Museum to permit easy access to the cultural heritage stored in museums around the world. The system provides a single virtual museum, enabling global search and edit of museum contents on the Internet. We applied the Global Digital Museum model to K-12 museum education by using real museum multimedia data. Technical issues addressed include: 1) unified and global access to heterogeneous and distributed multimedia contents of museums; and 2) interactive editing of the contents on the World-Wide Web. We describe the concept of Global Digital Museum, the system and network architecture, the data model for museum infomlation, and implementation of a prototype system on the Internet.