Gamma Modeling of Speech Power and Its On-Line Estimation for Statistical Speech Enhancement(Speech Enhancement, <Special Section> Statistical Modeling for Speech Processing)
スポンサーリンク
概要
- 論文の詳細を見る
This study shows the effectiveness of using gamma distribution in the speech power domain as a more general prior distribution for the model-based speech enhancement approaches. This model is a superset of the conventional Gaussian model of the complex spectrum and provides more accurate prior modeling when the optimal parameters are estimated. We develop a method to adapt the modeled distribution parameters from each actual noisy speech in a frame-by-frame manner. Next, we derive and investigate the minimum mean square error (MMSE) and maximum a posterior probability (MAP) estimations in different domains of speech spectral magnitude, generalized power and its logarithm, using the proposed gamma modeling. Finally, a comparative evaluation of the MAP and MMSE filters is conducted. As the MMSE estimations tend to more complicated using more general prior distributions, the MAP estimations are given in closed-form extractions and therefore are suitable in the implementation. The adaptive estimation of the modeled distribution parameters provides more accurate prior modeling and this is the principal merit of the proposed method and the reason for the better performance. From the experiments, the MAP estimation is recommended due to its high efficiency and low complexity. Among the MAP based systems, the estimation in log-magnitude domain is shown to be the best for the speech recognition as the estimation in power domain is superior for the noise reduction.
- 社団法人電子情報通信学会の論文
- 2006-03-01
著者
-
TAKEDA Kazuya
Nagoya University
-
Dat Tran
The Graduate School Of Information Science Nagoya University:presently With The Institute For Infoco
-
TAKEDA Kazuya
the Graduate School of Information Science, Nagoya University
-
ITAKURA Fumitada
the Graduate School of Information Engineering, Meijo University
-
Itakura Fumitada
The Graduate School Of Information Engineering Meijo University
-
Takeda Kazuya
The Graduate School Of Information Science Nagoya University
関連論文
- Acoustic Feature Transformation Combining Average and Maximum Classification Error Minimization Criteria
- Acoustic Feature Transformation Based on Discriminant Analysis Preserving Local Structure for Speech Recognition
- CENSREC-1-C : An evaluation framework for voice activity detection under noisy environments
- Driver Identification Using Driving Behavior Signals(Human-computer Interaction)
- AURORA-2J: An Evaluation Framework for Japanese Noisy Speech Recognition(Speech Corpora and Related Topics, Corpus-Based Speech Technologies)
- CENSREC-3: An Evaluation Framework for Japanese Speech Recognition in Real Car-Driving Environments(Speech and Hearing)
- Evaluation of HRTFs estimated using physical features
- Multiple Regression of Log Spectra for In-Car Speech Recognition Using Multiple Distributed Microphones(Feature Extraction and Acoustic Medelings, Corpus-Based Speech Technologies)
- Evaluation of Combinational Use of Discriminant Analysis-Based Acoustic Feature Transformation and Discriminative Training
- Acoustic Feature Transformation Based on Discriminant Analysis Preserving Local Structure for Speech Recognition
- Gamma Modeling of Speech Power and Its On-Line Estimation for Statistical Speech Enhancement(Speech Enhancement, Statistical Modeling for Speech Processing)
- Multichannel Speech Enhancement Based on Generalized Gamma Prior Distribution with Its Online Adaptive Estimation
- SNR and sub-band SNR estimation based on Gaussian mixture modeling in the log power domain with application for speech enhancements (第6回音声言語シンポジウム)
- SNR and sub-band SNR estimation based on Gaussian mixture modeling in the log power domain with application for speech enhancements (第6回音声言語シンポジウム)
- SNR and sub-band SNR estimation based on Gaussian mixture modeling in the log power domain with application for speech enhancements (第6回音声言語シンポジウム)
- Acoustic Feature Transformation Combining Average and Maximum Classification Error Minimization Criteria
- Driver's irritation detection using speech recognition results (音声・第10回音声言語シンポジウム)
- Driver's irritation detection using speech recognition results (音声言語情報処理)
- Driver's irritation detection using speech recognition results (言語理解とコミュニケーション・第10回音声言語シンポジウム)
- An Acoustically Oriented Vocal-Tract Model
- Estimation of speaker and listener positions in a car using binaural signals
- Sound localization under conditions of covered ears on the horizontal plane
- Single-Channel Multiple Regression for In-Car Speech Enhancement
- Adaptive Nonlinear Regression Using Multiple Distributed Microphones for In-Car Speech Recognition(Speech Enhancement, Multi-channel Acoustic Signal Processing)
- Speech Recognition Using Finger Tapping Timings(Speech and Hearing)
- CIAIR In-Car Speech Corpus : Influence of Driving Status(Corpus-Based Speech Technologies)
- Construction and Evaluation of a Large In-Car Speech Corpus(Speech Corpora and Related Topics, Corpus-Based Speech Technologies)
- Method for determining sound localization by auditory masking
- Acoustic Model Training Using Pseudo-Speaker Features Generated by MLLR Transformations for Robust Speaker-Independent Speech Recognition