Parameter Sharing in Mixture of Factor Analyzers for Speaker Identification(Feature Extraction and Acoustic Medelings, <Special Section>Corpus-Based Speech Technologies)
スポンサーリンク
概要
- 論文の詳細を見る
This paper investigates the parameter tying structures of a mixture of factor analyzers (MFA) and discriminative training of MFA for speaker identification. The parameters of factor loading matrices or diagonal matrices are shared in different mixtures of MFA. Then, minimum classification error (MCE) training is applied to the MFA parameters to enhance the discrimination ability. The result of a text-independent speaker identification experiment shows that MFA outperforms the conventional Gaussian mixture model (GMM) with diagonal or full covariance matrices and achieves the best performance when sharing the diagonal matrices, resulting in a relative gain of 26% over the GMM with diagonal covariance matrices. The improvement is more significant especially in sparse training data condition. The recognition performance is further improved by MCE training with an additional gain of 3% error reduction.
- 社団法人電子情報通信学会の論文
- 2005-03-01
著者
-
TOKUDA Keiichi
Department of Computer Science and Engineering, Nagoya Institute of Technology
-
Yamamoto Hiroyoshi
Department of Cardiology, National Hospital Organization Osaka National Hospital
-
Tokuda K
Department Of Computer Science And Engineering Nagoya Institute Of Technology
-
Tokuda Keiichi
Department Of Computer Science And Engineering Nagoya Institute Of Technology
-
NANKAKU Yoshihiko
Department of Computer Science and Engineering, Nagoya Institute of Technology
-
KITAMURA Tadashi
Department of Computer Science and Engineering, Nagoya Institute of Technology
-
MIYAJIMA Chiyomi
Department of Computer Science and Engineering, Nagoya Institute of Technology
-
Yamamoto Hiroyoshi
Department Of Computer Science And Engineering Nagoya Institute Of Technology
-
Yamamoto Hiroyoshi
Department Of Cardiology Osaka National Hospital
-
Miyajima Chiyomi
Department Of Media Science Nagoya University
-
Kitamura Tadashi
Department Of Computer Science And Engineering Nagoya Institute Of Technology
-
Kitamura Tadashi
Department Of Cardiothoracic Surgery The University Of Tokyo
-
Nankaku Yoshihiko
Department Of Computer Science And Engineering Nagoya Institute Of Technology
-
Tokuda Keiichi
Department Of Computer Science Naogya Institute Of Technology
-
Kitamura Tadashi
Department Of Cardiothoracic Surgery Faculty Of Medicine University Of Tokyo
-
Yamamoto Hiroyoshi
Department of Applied Chemistry, Faculty of Engineering, Osaka University
関連論文
- The Nitech-NAIST HMM-Based Speech Synthesis System for the Blizzard Challenge 2006
- Details of the Nitech HMM-Based Speech Synthesis System for the Blizzard Challenge 2005(Speech and Herring)
- Effects of Nicorandil on Cardiovascular Events in Patients With Coronary Artery Disease in The Japanese Coronary Artery Disease (JCAD) Study
- Gender Differences in Patients With Coronary Artery Disease in Japan : The Japanese Coronary Artery Disease Study (The JCAD Study)
- Beta-Blocker Prescription Among Japanese Cardiologists and Its Effect on Various Outcomes
- Relationship Between Renal Dysfunction and Severity of Coronary Artery Disease in Japanese Patients
- PJ-038 Cystatin C Predicts Severity of Coronary Artery Disease Even in Patients without Chronic Kidney Disease (CKD)(PJ007,Kidney/Renal Circulation/CKD 1 (H),Poster Session (Japanese),The 73rd Annual Scientific Meeting of The Japanese Circulation Society)
- PJ-483 Hemodynamic Effect Lasts for 2 Days after Short-term Infusion of Atrial Natriuretic Peptide in Patients with Chronic Heart Failure(Heart failure, clinical-12, The 71st Annual Scientific Meeting of the Japanese Circulation Society)
- PJ-057 Augmentation of Cardiac Parasympathetic Nerve Activity by Human Atrial Natriuretic Peptide and Its Mechanisms in Patients with Chronic Heart Failure(Heart failure, clinical-18 (M) PJ10,Poster Session (Japanese),The 70th Anniversary Annual Scientifi
- Natriuresis by Atrial Natriuretic Peptide is Related with the Decrease in Plasma Aldosterone Concentration in Patients with Acute Heart Failure (Heart Failure, Clinical 12 (M), The 69th Annual Scientific Meeting of the Japanese Circulation Society)
- Vagal Stimulation Decreases Plasma Aldosterone Concentration in Patients with Chronic Heart Failure(Heart Failure, Clinical 14 (M), The 69th Annual Scientific Meeting of the Japanese Circulation Society)
- PE-099 Correlation between pacing rate and daily physical activity in patients with VVI pacemaker implantation in atrial fibrillation(Arrhythmia, Non-Pharmacological Therapy 3 (A) : PE17)(Poster Session (English))
- Applying Sparse KPCA for Feature Extraction in Speech Recognition(Feature Extraction and Acoustic Medelings, Corpus-Based Speech Technologies)
- On the Use of Kernel PCA for Feature Extraction in Speech Recognition(Speech and Hearing)
- The Nitech-NAIST HMM-Based Speech Synthesis System for the Blizzard Challenge 2006
- A Hidden Semi-Markov Model-Based Speech Synthesis System(Speech and Hearing)
- State Duration Modeling for HMM-Based Speech Synthesis(Speech and Hearing)
- A Training Method of Average Voice Model for HMM-Based Speech Synthesis(Digital Signal Processing)
- A Context Clustering Technique for Average Voice Models (Special Issue on Speech Information Processing)
- Speaker Adaptation of Pitch and Spectrum for HMM-Based Speech Synthesis
- Multi-Space Probability Distribution HMM(Special Issue on the 2000 IEICE Excellent Paper Award)
- Vector Quantization of Speech Spectral Parameters Using Statistics of Static and Dynamic Features
- Text-Independent Speaker Identification Using Gaussian Mixture Models Based on Multi-Space Probability Distribution (Special Issue on Biometric Person Authentication)
- Establishment of a method of anonymization of DNA samples in genetic research
- A Reordering Model Using a Source-Side Parse-Tree for Statistical Machine Translation
- Clinical Experience with Cryopreserved Allografts for Aortic Infection
- A Fully Consistent Hidden Semi-Markov Model-Based Speech Recognition System
- Plasma Cystatin C Concentration Reflects the Severity of Coronary Artery Disease in Patients Without Chronic Kidney Disease
- Mixture Density Models Based on Mel-Cepstral Representation of Gaussian Process(Digital Signal Processing)
- A 16kb/s Wideband CELP-Based Speech Coder Using Mel-Generalized Cepstral Analysis
- Pseudoaneurysm Developed after Aortic Root Homograft Implantation
- Multiple Regression of Log Spectra for In-Car Speech Recognition Using Multiple Distributed Microphones(Feature Extraction and Acoustic Medelings, Corpus-Based Speech Technologies)
- LMS-Based Algorithms with Multi-Band Decomposition of the Estimation Error Applied to System Identification (Special Section on Digital Signal Processing)
- Multi-Band Decomposition of the Linear Prediction Error Applied to Adaptive AR Spectral Estimation
- Adaptive AR Spectral Estimation Based on Wavelet Decomposition of the Linear Prediction Error
- A Covariance-Typing Technique for HMM-Based Speech Synthesis
- Characteristics of Multi-Layer Perceptron Models in Enhancing Degraded Speech
- Prevalence of Vitreous Hemorrhage Following Coronary Revascularization in Patients With Diabetic Retinopathy
- Parameter Sharing in Mixture of Factor Analyzers for Speaker Identification(Feature Extraction and Acoustic Medelings, Corpus-Based Speech Technologies)
- Adaptive Nonlinear Regression Using Multiple Distributed Microphones for In-Car Speech Recognition(Speech Enhancement, Multi-channel Acoustic Signal Processing)
- Deterministic Annealing EM Algorithm in Acoustic Modeling for Speaker and Speech Recognition(Feature Extraction and Acoustic Medelings, Corpus-Based Speech Technologies)
- Continuous Speech Recognition Based on General Factor Dependent Acoustic Models(Feature Extraction and Acoustic Medelings, Corpus-Based Speech Technologies)
- Bayesian Context Clustering Using Cross Validation for Speech Recognition
- Physical Model-Based Indirect Measurement of Blood Flow and Pressures for Pulsatile Circulatory Assist-In Vitro Study
- Reformulating the HMM as a Trajectory Model
- Reformulating the HMM as a Trajectory Model
- Reformulating the HMM as a Trajectory Model
- Intensively Lowering Both Low-Density Lipoprotein Cholesterol and Blood Pressure Does Not Reduce Cardiovascular Risk in Japanese Coronary Artery Disease Patients
- ファジィ推論とリアルタイムモデルの併用について
- Speech recognition based on statistical models including multiple phonetic decision trees
- On the use of two-mass vocal cord model in characterizing the stress speech (音声)
- A Modeling Support Tool for a Global Human Model on the Internet
- The Development of a Physiological Simulation System for the Human Circulatory System Coupling Macro and Micro Models
- Report of the Committee on the classification and diagnostic criteria of diabetes mellitus : The Committee of the Japan Diabetes Society on the diagnostic criteria of diabetes mellitus
- Diagnostic Simulation Tool for a Circulatory System Model Based on Interpretive Structural Modeling
- How can a robot have consciousness?
- Animal-like Behavior Design of Small Robots by the Model of Subjective World and Behavior
- An Integrated Simulation Tool for Modeling the Human Circulatory System(Bioengineering)
- Animal-like behavior design of small robots by consciousness-based architecture
- A Case of Implantation of ICD over 30 Years after CABG for Coronary Arterial Lesions Due to Kawasaki Disease
- Moderate Prosthesis-Patient Mismatch May Be Negligible in Elderly Patients Undergoing Conventional Aortic Valve Replacement for Aortic Stenosis
- A Bayesian Framework Using Multiple Model Structures for Speech Recognition
- Neutrophil Elastase Inhibitor Sivelestat Attenuates Perioperative Inflammatory Response in Pediatric Heart Surgery With Cardiopulmonary Bypass:A Prospective Randomized Study
- Speaker interpolation for HMM-based speech synthesis system
- 1,3,5-Triarylpyrazoles as Phosphorescing Impurities in 1,3,5-Triaryl-2-prazolines