Hiroshi Shimodaira

Spoken Dialogue (2008)

Heriberto Cuayáhuitl, Supervisors Steve Renals, Oliver Lemon, Hiroshi Shimodaira, Hm M

Abstract: We propose an approach to learn reusable probabilistic models for simulating human-computer task-oriented dialogues at the intention level, trained from different corpora using a two-pass...

Sixth International Conference on Document Analysis and Recognition (ICDAR 2001), pp.491-495 (2001-09) Substroke Approach to HMM-based On-line Kanji Handwriting Recognition (2008)

Mitsuru Nakai, Naoto Akira, Hiroshi Shimodaira, Shigeki Sagayama

A new method is proposed for on-line handwriting recognition of Kanji characters. The method employs substroke HMMs as minimum units to constitute Japanese Kanji characters and utilizes the direction...

RESTORATION OF PITCH PATTERN OF SPEECH BASED ON A PITCH GENERATION MODEL (2008)

Hiroshi Shimodaira, Mitsuru Nakai, Akihiro Kumata

In this paper a model-based approach for restoring a continuous fundamental frequency (F0) contour from the noisy output of an F0 extractor is investigated. In contrast to the conventional pitch...

IMPROVING THE GENERALIZATION PERFORMANCE OF THE MCE/GPD LEARNING (2008)

Hiroshi Shimodaira, Jun Rokui, Mitsuru Nakai

A novel method to prevent the over-fitting effect and improve the generalization performance of the Minimum Classification Error (MCE) / Generalized Probabilistic Descent (GPD) learning is proposed....

On representation of fundamental frequency of speech for prosody analysis using reliability function,” EUROSPEECH’97 (2008)

Mitsuru Nakai, Hiroshi Shimodaira

This paper highlights on a method that provides a new prosodic feature called ‘F0 reliability field ’ based on a reliability function of the fundamental frequency (F0). The proposed method does...

2 (2008)

Shin-ichi Kawamoto, Hiroshi Shimodaira, Tsuneo Nitta, Satoshi Nakamura, Katsunobu Itou, Shigeo Morishima, ...

Summary. Galatea is a software toolkit to develop a human-like spoken dialog agent. In order to easily integrate the modules of different characteristics including speech recognizer, speech...

ACCENT PHRASE SEGMENTATION BY FINDING N-BEST SEQUENCES OF PITCH PATTERN TEMPLATES (2008)

Mitsuru Nakai, Hiroshi Shimodaira

This paper describes a prosodic method for segmenting continuous speech into accent phrases. Optimum sequences are obtained on the basis of least squared error criterion by using dynamic time warping...

THE USE OF F0 RELIABILITY FUNCTION FOR PROSODIC COMMAND ANALYSIS ON F0 CONTOUR GENERATION MODEL (2008)

Mitsuru Nakai, Hiroshi Shimodaira

This paper describes a method of utilizing an “F0 Reliability Field ” (FRF), which we have proposed in our previous work, for estimating prosodic commands on F0 contour generation model. This FRF...

— Draft of the extended abstract for MLMI2005 — Analysis and Synthesis of Head Motion for Lifelike Conversational Agents (2008)

Hiroshi Shimodaira, Keisuke Uematsu, Gregor Hofer, Mitsuru Nakai

This study aims to investigate which and what motions of lifelike conversational agents play essential role to make the agents natural. Some preliminary experimental results and future plan are...

A Divergent-Style Learning Support Tool for English Learners Using a Thesaurus Diagram (2008)

Chie Shimodaira, Hiroshi Shimodaira, Susumu Kunifuji

Abstract. This paper proposes an English learning support tool which provides users with divergent information to find the right words and expressions. In contrast to a number of software tools for...

8th International Workshop on Frontiers in Handwriting Recognition (IWFHR-8), pp.78-83 (2002-08) Context-dependent Substroke Model for HMM-based On-line Handwriting Recognition (2008)

Junko Tokuno, Nobuhito Inami, Shigeki Matsuda, Mitsuru Nakai, Hiroshi Shimodaira, Shigeki Sagayama

This paper describes context-dependent substroke hidden Markov models (HMMs) for on-line handwritten recognition of cursive Kanji and Hiragana characters. As there are more than 6,000 distinctive...

FEATURE-DEPENDENT ALLOPHONE CLUSTERING (2008)

Shigeki Matsuda, Mitsuru Nakai, Hiroshi Shimodaira, Shigeki Sagayama

We propose a novel method for clustering allophones called Feature-Dependent Allophone Clustering (FD-AC) that determines feature-dependent HMM topology automatically. Existing methods for allophone...

INTERSPEECH 2007 Hierarchical Dialogue Optimization Using Semi-Markov Decision Processes (2008)

Heriberto Cuayáhuitl, Steve Renals, Oliver Lemon, Hiroshi Shimodaira

This paper addresses the problem of dialogue optimization on large search spaces. For such a purpose, in this paper we propose to learn dialogue strategies using multiple Semi-Markov Decision...

16th International Conference on Pattern Recognition (ICPR 2002), Vol.III, pp.220-223 (2002-08) Pen Pressure Features for Writer-Independent On-line Handwriting Recognition Based on Substroke HMM (2008)

Mitsuru Nakai, Takashi Sudo, Hiroshi Shimodaira, Shigeki Sagayama

This paper discusses the use of pen pressure as a feature in writer-independent on-line handwriting recognition. We propose two kinds of features related to pen pressure: one is the pressure...

Modified Minimum Classification Error Learning and Its Application to Neural Networks (2008)

Hiroshi Shimodaira, Jun Rokui, Mitsuru Nakai

Abstract. A novel method to improve the generalization performance of the Minimum Classification Error (MCE) / Generalized Probabilistic Descent (GPD) learning is proposed. The MCE/GPD learning...

JACOBIAN ADAPTATION OF HMM WITH INITIAL MODEL SELECTION FOR NOISY SPEECH RECOGNITION (2008)

Hiroshi Shimodaira, Yutaka Kato, Toshihiko Akae, Mitsuru Nakai, Shigeki Sagayama

An extension of Jacobian Adaptation (JA) of HMMs for degraded speech recognition is presented in which appropriate set of initial models is selected from a number of initial-model sets designed for...

Hierarchical dialogue optimization using semi-markov decision processes. (2007)

Cuayáhuitl, Heriberto, Renals, Steve, Lemon, Oliver, Shimodaira, Hiroshi

This paper addresses the problem of dialogue optimization on large search spaces. For such a purpose, in this paper we propose to learn dialogue strategies using multiple Semi-Markov Decision...

Automatic Head Motion Prediction from Speech Data (2007)

Hofer, Gregor, Shimodaira, Hiroshi

In this paper we present a novel approach to generate a sequence of head motion units given some speech. The modelling approach is based on the notion that head motion can be divided into a number of...

Lip motion synthesis using a context dependent trajectory hidden Markov model (2007)

Hofer, Gregor, Shimodaira, Hiroshi, Yamagishi, Junichi

Lip synchronisation is essential to make character animation believeable. In this poster we present a novel technique to automatically synthesise lip motion trajectories given some text and speech....

Speech-driven head motion synthesis based on a trajectory model. (2007)

Hofer, Gregor, Shimodaira, Hiroshi, Yamagishi, Junichi

Making human-like characters more natural and life-like requires more inventive approaches than current standard techniques such as synthesis using text features or triggers. In this poster we...

Hierarchical dialogue optimization using semi-markov decision processes (2007)

Heriberto Cuayáhuitl, Steve Renals, Oliver Lemon, Hiroshi Shimodaira

This paper addresses the problem of dialogue optimization on large search spaces. For such a purpose, in this paper we propose to learn dialogue strategies using multiple Semi-Markov Decision...

On-line Handwritten Character Recognition Selectively Employing Hierarchical Spatial Relationships among Subpatterns (2006)

Tokuno, Junko, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki, Nakagawa, Masaki

This paper proposes an on-line handwritten character pattern recognition method that examines spatial relationships among subpatterns which are components of a character pattern. Conventional methods...

On-line Handwritten Character Recognition Selectively Employing Hierarchical Spatial Relationships among Subpatterns (2006)

Tokuno, Junko, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki, Nakagawa, Masaki

This paper proposes an on-line handwritten character pattern recognition method that examines spatial relationships among subpatterns which are components of a character pattern. Conventional methods...

On-line Handwritten Character Recognition Selectively Employing Hierarchical Spatial Relationships among Subpatterns (2006)

Tokuno, Junko, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki, Nakagawa, Masaki

This paper proposes an on-line handwritten character pattern recognition method that examines spatial relationships among subpatterns which are components of a character pattern. Conventional methods...

Learning Multi-Goal Dialogue Strategies Using (2006)

Reinforcement Learning With, Heriberto Cuayáhuitl, Steve Renals, Oliver Lemon, Hiroshi Shimodaira

Learning dialogue strategies using the reinforcement learning framework is problematic due to its expensive computational cost. In this paper we propose an algorithm that reduces a state-action space...

Reinforcement Learning Of Dialogue Strategies (2006)

With Hierarchic Al, Heriberto Cuayáhuitl, Steve Renals, Oliver Lemon, Hiroshi Shimodaira

MACHINES Heriberto Cuay ahuitl , Steve Renals , Oliver Lemon , Hiroshi Shimodaira CSTR , HCRC , School of Informatics, University of Edinburgh 2 Buccleuch Place, EH8 9LW, Edinburgh, Scotland, UK...

Learning Multi-Goal Dialogue Strategies Using Reinforcement Learning With Reduced State-Action Spaces (2006)

Heriberto Cuayáhuitl, Steve Renals, Oliver Lemon, Hiroshi Shimodaira

Learning dialogue strategies using the reinforcement learning framework is problematic due to its expensive computational cost. In this paper we propose an algorithm that reduces a state-action space...

Analysis and Synthesis of Head Motionfor Life like Conversational Agents (2005)

Shimodaira, Hiroshi, Uematsu, Keisuke, Kawamoto, Shin Ichi, Hofer, Gregor O, Nakai, Mitsuru

This study aims to investigate which and what motions of lifelike conversational agents play essential role to make the agents natural. Some preliminary experimental results and future plan are...

Analysis and Synthesis of Head Motionfor Life like Conversational Agents (2005)

Shimodaira, Hiroshi, Uematsu, Keisuke, Kawamoto, Shin Ichi, Hofer, Gregor O, Nakai, Mitsuru

This study aims to investigate which and what motions of lifelike conversational agents play essential role to make the agents natural. Some preliminary experimental results and future plan are...

Human-Computer Dialogue Simulation Using Hidden Markov Models (2005)

Heriberto Cuay Ahuitl, Heriberto Cuayáhuitl, Steve Renals, Oliver Lemon, Hiroshi Shimodaira

This paper presents a probabilistic method to simulate task-oriented human-computer dialogues at the intention level, that may be used to improve or to evaluate the performance of spoken dialogue...

Learning Reusable Probabilistic User-System Models for Human-Computer Dialogue Simulation (2004)

Heriberto Cuayáhuitl, Supervisors Steve Renals, Oliver Lemon, Hiroshi Shimodaira, Hm M

We propose an approach to learn reusable probabilistic models for simulating human-computer task-oriented dialogues at the intention level, trained from di#erent corpora using a two-pass clustering...

On-line Overlaid-Handwriting Recognition Based on Substroke HMMs. (2003)

Shimodaira, Hiroshi, Sudo, Takashi, Nakai, Mitsuru, Sagayama, Shigeki

This study discusses the subject of training data selection for neural networks using back propagation. We have made only one assumption that there are no overlapping of training data belonging to...

Life-Like Characters. Tools, Affective Functions, and Applications (2003)

Kawamoto, Shin Ichi, Shimodaira, Hiroshi, Nitta, Tsuneo, Nishimoto, Takuya, Nakamura, Satoshi, Itou, Katsunobu, ...

Galatea is a software toolkit to develop a human-like spoken dialog agent. In order to easily integrate the modules of different characteristics including speech recognizer, speech synthesizer,...

Generation of Hierarchical Dictionary for Stroke-order Free Kanji Handwriting Recognition Based on Substroke HMM. (2003)

Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

This paper describes a method of generating a Kanji hierarchical structured dictionary for stroke-number and stroke-order free handwriting recognition based on sub-stroke HMM. In stroke-based...

On-line Overlaid-Handwriting Recognition Based on Substroke HMMs. (2003)

Shimodaira, Hiroshi, Sudo, Takashi, Nakai, Mitsuru, Sagayama, Shigeki

This study discusses the subject of training data selection for neural networks using back propagation. We have made only one assumption that there are no overlapping of training data belonging to...

Life-Like Characters. Tools, Affective Functions, and Applications (2003)

Kawamoto, Shin Ichi, Shimodaira, Hiroshi, Nitta, Tsuneo, Nishimoto, Takuya, Nakamura, Satoshi, Itou, Katsunobu, ...

Galatea is a software toolkit to develop a human-like spoken dialog agent. In order to easily integrate the modules of different characteristics including speech recognizer, speech synthesizer,...

Generation of Hierarchical Dictionary for Stroke-order Free Kanji Handwriting Recognition Based on Substroke HMM (2003)

Mitsuru Nakai Hiroshi, Hiroshi Shimodaira, Shigeki Sagayama

This paper describes a method of generating a Kanji hierarchical structured dictionary for stroke-number and stroke-order free handwriting recognition based on substroke HMM. In stroke-based methods,...

Generation of Hierarchical Dictionary for Stroke-order Free Kanji Handwriting Recognition Based on Substroke HMM (2003)

Mitsuru Nakai, Hiroshi Shimodaira, Shigeki Sagayama

This paper describes a method of generating a Kanji hierarchical structured dictionary for stroke-number and stroke-order free handwriting recognition based on substroke HMM. In stroke-based methods,...

Online Overlaid-Handwriting Recognition Based on Substroke HMMs (2003)

Hiroshi Shimodaira, Takashi Sudo, Mitsuru Nakai, Shigeki Sagayama

This paper proposes a novel handwriting recognition interface for wearable computing where users write characters continuously without pauses on a small single writing box. Since characters are...

Hidden Markov Model for Automatic Transcription of MIDI Signals (2002)

Takeda, Haruto, Saito, Naoki, Otsuki, Tomoshi, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

This paper describes a Hidden Markov Model (HMM)-based method of automatic transcription of MIDI (Musical Instrument Digital Interface) signals of performed music. The problem is formulated as...

Hidden Markov Model for Automatic Transcription of MIDI Signals (2002)

Takeda, Haruto, Saito, Naoki, Otsuki, Tomoshi, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

This paper describes a Hidden Markov Model (HMM)-based method of automatic transcription of MIDI (Musical Instrument Digital Interface) signals of performed music. The problem is formulated as...

Pen pressure features for writer-independent on-line handwriting recognition based on substroke HMM (2002)

Nakai, Mitsuru, Sudo, Takashi, Shimodaira, Hiroshi, Sagayama, Shigeki

This paper discusses the use of pen pressure as a feature in writer-independent on-line handwriting recognition. We propose two kinds of features related to pen pressure: one is the pressure...

Context-dependent substroke model for HMM-based on-line handwriting recognition (2002)

Tokuno, Junko, Inami, Nobuhito, Matsuda, Shigeki, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

Describes context-dependent substroke hidden Markov models (HMMs)for on-line handwritten recognition of cursive Kanji and Hiragana characters. In order to tackle this problem, we have proposed the...

Pen pressure features for writer-independent on-line handwriting recognition based on substroke HMM (2002)

Nakai, Mitsuru, Sudo, Takashi, Shimodaira, Hiroshi, Sagayama, Shigeki

This paper discusses the use of pen pressure as a feature in writer-independent on-line handwriting recognition. We propose two kinds of features related to pen pressure: one is the pressure...

Context-dependent substroke model for HMM-based on-line handwriting recognition (2002)

Tokuno, Junko, Inami, Nobuhito, Matsuda, Shigeki, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

Describes context-dependent substroke hidden Markov models (HMMs)for on-line handwritten recognition of cursive Kanji and Hiragana characters. In order to tackle this problem, we have proposed the...

Jacobian Joint Adaptation to Noise, Channel and Vocal Tract Length (2002)

Shimodaira, Hiroshi, Sakai, Nobuyoshi, Nakai, Mitsuru, Sagayama, Shigeki

A new Jacobian approach that linearly decomposes the composite of additive noise, multiplicative noise (channel transfer function) and speaker's vocal tract length, and adapts the acoustic model...

Jacobian Joint Adaptation to Noise, Channel and Vocal Tract Length (2002)

Shimodaira, Hiroshi, Sakai, Nobuyoshi, Nakai, Mitsuru, Sagayama, Shigeki

A new Jacobian approach that linearly decomposes the composite of additive noise, multiplicative noise (channel transfer function) and speaker's vocal tract length, and adapts the acoustic model...

Open-source Software for Developing Anthropomorphic Spoken Dialog Agents (2002)

Shin-ichi, Kawamoto, Shimodaira, Hiroshi, Nitta, Tsuneo, Nishimoto, Takuya, Nakamura, Satoshi, Itou, Katsunobu, ...

An architecture for highly-interactive human-like spoken-dialog agent is discussed in this paper. In order to easily integrate the modules of different characteristics including speech recognizer,...

Open-source Software for Developing Anthropomorphic Spoken Dialog Agents (2002)

Shin-ichi, Kawamoto, Shimodaira, Hiroshi, Nitta, Tsuneo, Nishimoto, Takuya, Nakamura, Satoshi, Itou, Katsunobu, ...

An architecture for highly-interactive human-like spoken-dialog agent is discussed in this paper. In order to easily integrate the modules of different characteristics including speech recognizer,...

Hidden Markov Model for Automatic Transcription (2002)

Haruto Takeda, Naoki Saito, Mitsuru Nakai, Hiroshi Shimodaira, Tomoshi Otsuki, Shigeki Sagayama

mit,sim£ Abstract — This paper describes a Hidden Markov Model (HMM)-based method of automatic transcription of MIDI (Musical Instrument Digital Interface) signals of performed music. The problem...

Jacobian joint adaptation to noise, channel and vocal tract length (2002)

Hiroshi Shimodaira, Nobuyoshi Sakai, Mitsuru Nakai

A new Jacobian approach that linearly decomposes the composite of additive noise, multiplicative noise (channel transfer function) and speaker’s vocal tract length, and adapts the acoustic model...

Multiple-Regression Hidden Markov Model (2001)

Fujinaga, Katsuhisa, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

This paper proposes a new class of hidden Markov model (HMM) called multiple-regression HMM (MRHMM) that utilizes auxiliary features such as fundamental frequency (F0) and speaking styles that affect...

Substroke Approach to HMM-based On-line Kanji Handwriting Recognition. (2001)

Nakai, Mitsuru, Akira, Naoto, Shimodaira, Hiroshi, Sagayama, Shigeki

A new method is proposed for on-line handwriting recognition of Kanji characters. The method employs substroke HMMs as minimum units to constitute Japanese Kanji characters and utilizes the direction...

Dynamic Time-Alignment Kernel in Support Vector Machine. (2001)

Shimodaira, Hiroshi, Noma, Ken Ichi, Nakai, Mitsuru, Sagayama, Shigeki

A new class of Support Vector Machine (SVM) that is applicable to sequential-pattern recognition such as speech recognition is developed by incorporating an idea of non-linear time alignment into the...

Multiple-Regression Hidden Markov Model (2001)

Fujinaga, Katsuhisa, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

This paper proposes a new class of hidden Markov model (HMM) called multiple-regression HMM (MRHMM) that utilizes auxiliary features such as fundamental frequency (F0) and speaking styles that affect...

Substroke Approach to HMM-based On-line Kanji Handwriting Recognition. (2001)

Nakai, Mitsuru, Akira, Naoto, Shimodaira, Hiroshi, Sagayama, Shigeki

A new method is proposed for on-line handwriting recognition of Kanji characters. The method employs substroke HMMs as minimum units to constitute Japanese Kanji characters and utilizes the direction...

Dynamic Time-Alignment Kernel in Support Vector Machine. (2001)

Shimodaira, Hiroshi, Noma, Ken Ichi, Nakai, Mitsuru, Sagayama, Shigeki

A new class of Support Vector Machine (SVM) that is applicable to sequential-pattern recognition such as speech recognition is developed by incorporating an idea of non-linear time alignment into the...

Multiple-Regression Hidden Markov Model (2001)

Katsuhisa Fujinaga, Mitsuru Nakai, Hiroshi Shimodaira, Shigeki Sagayama

This paper proposes a new class of hidden Markov model (HMM) called multiple-regression HMM (MRHMM) that utilizes auxiliary features such as fundamental frequency ( ) and speaking styles that affect...

Jacobian Adaptation of HMM with Initial Model Selection for Noisy Speech Recognition (2000)

Shimodaira, Hiroshi, Kato, Yukata, Akae, Toshihiko, Nakai, Mitsuru, Sagayama, Shigeki

An extension of Jacobian Adaptation (JA) of HMMs for degraded speech recognition is presented in which appropriate set of initial models is selected from a number of initial-model sets designed for...

Feature-Dependent Allophone Clustering (2000)

Matsuda, Shigeki, Naiki, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

We propose a novel method for clustering allophones called Feature-Dependent Allophone Clustering (FD-AC) that determines feature-dependent HMM topology automatically. Existing methods for allophone...

Jacobian Adaptation of HMM with Initial Model Selection for Noisy Speech Recognition (2000)

Shimodaira, Hiroshi, Kato, Yukata, Akae, Toshihiko, Nakai, Mitsuru, Sagayama, Shigeki

An extension of Jacobian Adaptation (JA) of HMMs for degraded speech recognition is presented in which appropriate set of initial models is selected from a number of initial-model sets designed for...

Feature-Dependent Allophone Clustering (2000)

Matsuda, Shigeki, Naiki, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

We propose a novel method for clustering allophones called Feature-Dependent Allophone Clustering (FD-AC) that determines feature-dependent HMM topology automatically. Existing methods for allophone...

Asynchronous-Transition HMM (2000)

Matsuda, Shigeki, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

We propose a new class of hidden Markov model (HMM) called asynchronous-transition HMM (AT-HMM). Opposed to conventional HMMs where hidden state transition occurs simultaneously to all features, the...

Asynchronous-Transition HMM (2000)

Matsuda, Shigeki, Nakai, Mitsuru, Shimodaira, Hiroshi, Sagayama, Shigeki

We propose a new class of hidden Markov model (HMM) called asynchronous-transition HMM (AT-HMM). Opposed to conventional HMMs where hidden state transition occurs simultaneously to all features, the...

A Training Scheme for Pattern Classification Using Multi-layer Feed-forward Neural Networks. (1999)

Keeni, Kanad, Nakayama, Kenji, Shimodaira, Hiroshi

This study highlights on the subject of weight initialization in multi-layer feed-forward networks. Training data is analyzed and the notion of criti- cal point is introduced for determining the...

A Training Scheme for Pattern Classification Using Multi-layer Feed-forward Neural Networks. (1999)

Keeni, Kanad, Nakayama, Kenji, Shimodaira, Hiroshi

This study highlights on the subject of weight initialization in multi-layer feed-forward networks. Training data is analyzed and the notion of criti- cal point is introduced for determining the...

Asynchronous-Transition HMM for Acoustic Modeling (1999)

Shigeki Sagayama, Shigeki Matsuda, Mitsuru Nakai, Hiroshi Shimodaira

We propose a new class of hidden Markov model (HMM) which we call Asynchronous-Transition HMM (AT-HMM) to model asynchronous temporal structure of acoustic feature sequences. Conventional HMM models...

Improving The Generalization Performance Of The MCE/GPD Learning (1998)

Shimodaira, Hiroshi, Rokui, Jun, Nakai, Mitsuru

A novel method to prevent the over-fitting effect and improve the generalization performance of the Minimum Classification Error (MCE) / Generalized Probabilistic Descent (GPD) learning is proposed....

The Use of F0 Reliability Function for Prosodic Command Analysis on F0 Contour Generation Model (1998)

Nakai, Mitsuru, Shimodaira, Hiroshi

This paper describes a method of utilizing an ``F0 Reliability Field'' (FRF), which we have proposed in our previous work, for estimating prosodic commands on F0 contour generation model. This FRF is...

Improving The Generalization Performance Of The MCE/GPD Learning (1998)

Shimodaira, Hiroshi, Rokui, Jun, Nakai, Mitsuru

A novel method to prevent the over-fitting effect and improve the generalization performance of the Minimum Classification Error (MCE) / Generalized Probabilistic Descent (GPD) learning is proposed....

The Use of F0 Reliability Function for Prosodic Command Analysis on F0 Contour Generation Model (1998)

Nakai, Mitsuru, Shimodaira, Hiroshi

This paper describes a method of utilizing an ``F0 Reliability Field'' (FRF), which we have proposed in our previous work, for estimating prosodic commands on F0 contour generation model. This FRF is...

Automatic Generation of Initial Weights and Estimation of Hidden Units for Pattern Classifcation Using Neural Networks (1998)

Keeni, Kanad, Nakayama, Kenji, Shimodaira, Hiroshi

This study high lights on the subject of weight initialization in back-propagation feed-forward networks. Training data is analyzed and the notion of critical points is introduced for determining the...

Modifed Minimum Classification Error Learning and Its Application to Neural Networks (1998)

Shimodaira, Hiroshi, Rokui, Jun, Nakai, Mitsuru

A novel method to improve the generalization performance of the Minimum Classification Error (MCE) / Generalized Probabilistic Descent (GPD) learning is proposed. The MCE/GPD learning proposed by...

Automatic Generation of Initial Weights and Estimation of Hidden Units for Pattern Classifcation Using Neural Networks (1998)

Keeni, Kanad, Nakayama, Kenji, Shimodaira, Hiroshi

This study high lights on the subject of weight initialization in back-propagation feed-forward networks. Training data is analyzed and the notion of critical points is introduced for determining the...

Modifed Minimum Classification Error Learning and Its Application to Neural Networks (1998)

Shimodaira, Hiroshi, Rokui, Jun, Nakai, Mitsuru

A novel method to improve the generalization performance of the Minimum Classification Error (MCE) / Generalized Probabilistic Descent (GPD) learning is proposed. The MCE/GPD learning proposed by...

Restration of Pitch Pattern of Speech Based on a Pitch Gereration Model. (1997)

Shimodaira, Hiroshi, Nakai, Mitsuru, Kumata, Akihiro

In this paper a model-based approach for restoring a continuous fundamental frequency (F0) contour from the noisy output of an F0 extractor is investigated. In contrast to the conventional pitch...

On Representation of Fundamental Frequency of Speech for Prosody Analysis Using Reliability Function. (1997)

Nakai, Mitsuru, Shimodaira, Hiroshi

This paper highlights on a method that provides a new prosodic feature called ‘F0 reliability field’ based on a reliability function of the fundamental frequency (F0). The proposed method does...

Restration of Pitch Pattern of Speech Based on a Pitch Gereration Model. (1997)

Shimodaira, Hiroshi, Nakai, Mitsuru, Kumata, Akihiro

In this paper a model-based approach for restoring a continuous fundamental frequency (F0) contour from the noisy output of an F0 extractor is investigated. In contrast to the conventional pitch...

On Representation of Fundamental Frequency of Speech for Prosody Analysis Using Reliability Function. (1997)

Nakai, Mitsuru, Shimodaira, Hiroshi

This paper highlights on a method that provides a new prosodic feature called ‘F0 reliability field’ based on a reliability function of the fundamental frequency (F0). The proposed method does...

Using Prosodic Information to Constrain Language Models for Spoken Dialogue (1996)

Taylor, Paul A, Shimodaira, Hiroshi, Isard, Stephen, King, Simon, Kowtko, Jacqueline C

We present work intended to improve speech recognition performance for computer dialogue by taking into account the way that dialogue context and intonational tune interact to limit the possibilities...

Using Prosodic Information to Constrain Language Models for Spoken Dialogue (1996)

Taylor, Paul A, Shimodaira, Hiroshi, Isard, Stephen, King, Simon, Kowtko, Jacqueline C

We present work intended to improve speech recognition performance for computer dialogue by taking into account the way that dialogue context and intonational tune interact to limit the possibilities...

Using prosodic information to constrain language models for spoken dialogue (1996)

Paul Taylor, Hiroshi Shimodaira, Stephen Isard, Simon King, Jaqueline Kowtko

We present work intended to improve speech recognition performance for computer dialogue by taking into account the way that dialogue context and intonational tune interact to limit the possibilities...

Using Prosodic Information to Constrain Language Models for Spoken Dialogue (1996)

Paul Taylor, Hiroshi Shimodaira, Stephen Isard, Simon King, Jaqueline Kowtko

We present work intended to improve speech recognition performance for computer dialogue by taking into account the way that dialogue context and intonational tune interact to limit the possibilities...

Automatic Prosodic Segmentation by F0 Clustering Using Superpositional Modeling. (1995)

Nakai, Mitsuru, Harald, Singer, Sagisaka, Yoshinori, Shimodaira, Hiroshi

In this paper, we propose an automatic method for detecting accent phrase boundaries in Japanese continuous speech by using F0 information. In the training phase, hand labeled accent patterns are...

Automatic Prosodic Segmentation by F0 Clustering Using Superpositional Modeling. (1995)

Nakai, Mitsuru, Harald, Singer, Sagisaka, Yoshinori, Shimodaira, Hiroshi

In this paper, we propose an automatic method for detecting accent phrase boundaries in Japanese continuous speech by using F0 information. In the training phase, hand labeled accent patterns are...

Automatic Prosodic Segmentation by F0 Clustering Using Superpositional Modeling (1995)

Mitsuru Nakai, Harald Singer, Yoshinori Sagisaka, Hiroshi Shimodaira

In this paper, we propose an automatic method for detecting accent phrase boundaries in Japanese continuous speech by using F0 information. In the training phase, hand labeled accent patterns are...

Accent Phrase Segmentation by Finding N-Best Sequences of Pitch Pattern Templates (1994)

Nakai, Mitsuru, Shimodaira, Hiroshi

This paper describes a prosodic method for segmenting continuous speech into accent phrases. Optimum sequences are obtained on the basis of least squared error criterion by using dynamic time warping...

Accent Phrase Segmentation by Finding N-Best Sequences of Pitch Pattern Templates (1994)

Nakai, Mitsuru, Shimodaira, Hiroshi

This paper describes a prosodic method for segmenting continuous speech into accent phrases. Optimum sequences are obtained on the basis of least squared error criterion by using dynamic time warping...

Prosodic phrase segmentation by pitch pattern clustering (1994)

Shimodaira, Hiroshi, Nakai, Mitsuru

This paper proposes a novel method for detecting the optimal sequence of prosodic phrases from continuous speech based on data-driven approach. The pitch pattern of input speech is divided into...

Prosodic phrase segmentation by pitch pattern clustering (1994)

Shimodaira, Hiroshi, Nakai, Mitsuru

This paper proposes a novel method for detecting the optimal sequence of prosodic phrases from continuous speech based on data-driven approach. The pitch pattern of input speech is divided into...

Prosodic phrase segmentation by pitch pattern clustering (1994)

Hiroshi Shimodaira, Mitsuru Nakai

This paper proposes a novel method for detecting the optimal sequence of prosodic phrases from continuous speech based on data-driven approach. The pitch pattern of input speech is divided into...

Accent phrase segmentation using transition probabilities between pitch pattern templates. (1993)

Shimodaira, Hiroshi, Nakai, Mitsuru

This paper proposes a novel method for segmenting continuous speech into accent phrases by using a prosodic feature 'pitch pattern'. The pitch pattern extracted from input speech signals is divided...

Accent phrase segmentation using transition probabilities between pitch pattern templates. (1993)

Shimodaira, Hiroshi, Nakai, Mitsuru

This paper proposes a novel method for segmenting continuous speech into accent phrases by using a prosodic feature 'pitch pattern'. The pitch pattern extracted from input speech signals is divided...

Robust Pitch Detection by Narrow Band Spectrum Analysis (1992)

Shimodaira, Hiroshi, Nakai, Mitsuru

This paper proposes a new technique for detecting pitch patterns which is useful for automatic speech recognition, by using a narrow band spectrum analysis. The motivation of this approach is that...

Robust Pitch Detection by Narrow Band Spectrum Analysis (1992)

Shimodaira, Hiroshi, Nakai, Mitsuru

This paper proposes a new technique for detecting pitch patterns which is useful for automatic speech recognition, by using a narrow band spectrum analysis. The motivation of this approach is that...

Robust Pitch Detection by Narrow Band Spectrum Analysis (1992)

Hiroshi Shimodaira, Mitsuru Nakai

This paper proposes a new technique for detecting pitch patterns which is useful for automatic speech recognition, by using a narrow band spectrum analysis. The motivation of this approach is that...