Shuji Taniguchi

Recurrent neural networks for phoneme recognition

This paper deals with recurrent neural networks of multilayer perceptron type which are well-suited for speech recognition, specially for phoneme recognition. The ability of these networks has been investigated by phoneme recognition experiments using a number of Japanese words uttered by a native male speaker in a quiet environment. Results of the experiments show that recognition rates achieved with these networks are higher than those obtained with conventional non-recurrent neural networks.

A path planning algorithm for multi manipulators

This paper discusses an approach to the development of a path planning algorithm for cooperating multi manipulators. In the approach, multiple manipulators are considered as a single composite one. Therefore, the composite manipulator possesses all arms of the manipulators included in the collaboration. A newly developed path planning algorithm BFA (Backtrack Free path planning Algorithm) enables the efficient generation of paths of this composite manipulator with extremely many arms. The algorithm is backtrack free and resolution complete. Computation volume of the algorithm is proportional to the total number of arms included in the composite manipulator. An additional advantage of this approach is that paths of individual manipulators can be calculated completely in parallel.

Feasiblity of Hungarian algorithm based Scheduling

An optimal resource allocation algorithm, Hungarian algorithm, is not directly applicable to manufacturing scheduling problems, because solutions of resource allocation problems may violate precedence constraints among processes that constitute individual manufacturing jobs. To apply Hungarian algorithm to scheduling problems, in this paper, several strategies for assigning prices to time slots of individual machines, which are allocated to processes, are proposed. Preliminary experimentation results showed that these strategies can generate near optimal schedules, i.e. when lengths of scheduling horizons were larger than 3 times of the maximum lengths of jobs, generated schedules could complete given jobs while maintaining the deterioration of the efficiency less than 5% from optimal schedules.

Development of Anonymous Networks Based on Symmetric Key Encryptions

Anonymous networks enable entities to send messages without disclosing their identities. Many anonymous networks had been proposed already, such as Mixnet, DC-net, Crowds, etc., however, they still have serious drawbacks. Namely, they require tremendous computation overheads to transmit messages over networks. That is because asymmetric key encryption algorithms are used. This paper proposes ESEBM (Enhanced Symmetric Key Encryption based Mixnet), a new mechanism for anonymous communication that removes drawbacks of existing anonymous networks while exploiting symmetric key encryption algorithms. According to experimentations, throughput of ESEBM is about 1/4.4 of usual non-anonymous networks, and it achieves more than 36 times higher throughput compared with Mixnet. In addition, different from existing anonymous networks, ESEBM can handle reply messages without any additional mechanism, and it can protect itself from various threats, e.g. DOS attacks and message forgeries.

Speech recognition based on a model of human auditory system

The paper deals with a new phoneme recognition system based on a model of human auditory system. This system is made up of a model of the human cochlea and a simple multilayer recurrent neural network which has feedback connections of self-loop type. The ability of this system has been investigated by a phoneme recognition experiment using a number of Japanese words uttered by a native male speaker. The result of the experiment shows that recognition accuracies achieved with this system in the presence of noise are higher than those obtained by a combination of frequency spectral analysis by DFT and a conventional feedforward neural network and that the cochlea model effectively prevents deterioration due to noise of recognition accuracy.

A scheme for collecting anonymous data

This paper proposes a scheme for calculating aggregate values of data owned by anonymous entities. Linear mix-nets, an unknown unique number generator, and anonymous tag based credentials efficiently conceal not only identities of data owners but also linkages between data owned by same entities. Then, the scheme enables quick introduction of advancing information technologies to industrial applications while ensuring confidentiality of sensitive data, e.g. a machine maintenance company can monitor states of machines remotely without knowing secrets of machine owners, companies can outsource even their sensitive tasks without worrying about leaks of their secrets. The scheme is also applicable to tasks in e-governance systems such as tax-collections where privacies of individuals must be preserved.

A Mechanism to Protect Tokens from Dishonest Uses

Off-line anonymous credit card systems are ones that enable card holders to make purchases anonymously while being disconnected from a card company, i.e. without using information maintained the by card company or other shops. This paper proposes one of components that are required for developing off-line anonymous credit card systems. The proposed component enables card holders to protect their tokens from being used by other entities.

Kana-to-kanji conversion method using Markov chain model of words in bunsetsu

We previously proposed a kana-to-kanji conversion method of non-segmented kana sentences by using Markov chain model of words in sentence. However, we could not obtain the enough accuracy rate for conversion by this method. The cause is considered that the total number of the rules is not saturated in the dictionary of Markov chain probabilities of words in sentence. Therefore, we take notice that the total number of the rules is almost saturated in the dictionary of Markov chain probabilities of words in bunsetsu. In this paper, we propose a new kana-to-kanji conversion method by using this Markov chain model. That is, the new proposed method detects simultaneously the boundaries of kana bunsetsu in sentence and the boundaries of kana word in bunsetsu by using Markov chain model of kana words in bunsetsu, and then converts kana words to the candidates of kanji-kana word and selects the maximum likely candidate by using Markov chain model of kanji-kana words in bunsetsu. Through the experiments by using statistical data of daily Japanese newspaper, the previous proposed method (called Method-B1) and the new proposed method (called Method-B2) are evaluated by the criteria of the accuracy rate for conversion. From the results of the experiments, it is concluded that Method-B2 is superior to Method-B1 in the accuracy rate for conversion and is effective in kana-to-kanji conversion of non-segmented kana sentences.

Speech recognition based on subword units

Large vocabulary, isolated word recognition requires a large amount of training data proportional to the vocabulary size to characterize each individual word model. A subword‐unit‐based approach is a more viable alternative than the word‐based approach to overcome the problem of the training data size, since different words can share common segments in their representations in the former. This paper deals with a couple of isolated word recognition systems where the subword‐unit‐based approach is commonly employed, though their methods of segmentation are completely different. In one system a hidden Markov model is used to decompose a word into subword units (segments), and frequency spectra of those subword units are fed to a recurrent neural network to yield a subword code sequence for the word. This sequence is then recognized hopefully as the original word by a set of hidden Markov models for isolated words. In the other system subword boundaries within a word are detected by finding peaks of the delta...

A New Method to Reduce the Ambiguity of Japanese Phoneme Candidates Recognized by Recurrent Neural Networks

Up to now, the method to reduce the ambiguity of phoneme recognition using 2nd-order Markov chain model of phonemes, has been proposed and has been evaluated by phonem lattice simulated and limited to substitution error. However, the method will be necessary to demonstrate the effectiveness for the phoneme candidate lattice obtained by actual speech recognition devices. This paper deals with recurrent neural networks(RNN) which are well- suited for natural language processing of speech recognition, specially for phoneme recognition. The ability of these networks has been investigated by phoneme recognition experiments using a number of Japanese words uttered by a native male speaker in a quiet environment. A method to detect the locations of devoicing vowels using the short- time average energy has been also proposed, and evaluated. Form results of the experiments, it is shown that recognition rates achieved with RNN are higher than those obtained with conventional non-recurrent neural networks, and that the method to detect the locations of devoicing vowels is useful.