Improved Global Soft Decision Using Smoothed Global Likelihood Ratio for Speech Enhancement(Multimedia Systems for Communications)
スポンサーリンク
概要
- 論文の詳細を見る
In this letter, we propose an improved global soft decision for noisy speech enhancement. From an investigation of statistical model-based speech enhancement, it is discovered that a global soft decision has a fundamental drawback at the speech tail regions of speech signals. For that reason, we propose a new solution based on a smoothed likelihood ratio for the global soft decision. Performances of the proposed method are evaluated by subjective tests under various environments and show better results compared with the our previous work.
- 社団法人電子情報通信学会の論文
- 2007-08-01
著者
-
Jeong Dong
School Of Electronic Engineering Inha University
-
CHANG Joon-Hyuk
School of Electronic Engineering, Inha University
-
Chang Joon-hyuk
School Of Electronic Engineering Inha University
-
Chang Joon-hyuk
School Of Electronic And Electrical Engineering Inha University
-
KIM Nam
Institute for Laser Science, Univ.of Electro-Communications
-
Kim Nam
Institute Of New Media And Communications School Of Electrical Engineering And Computer Science Seou
-
KANG Sangki
Telecommuniation R & D Center, Samsung Electronics
-
Kang Sangki
Telecommuniation R & D Center Samsung Electronics
-
Kim Nam
Institute For Clinical Research College Of Medicine Pochon Cha University
-
Jeong Dong
School of Display Engineering, Hoseo University, Baebang-myun, Asan, Chungnam 336-795, Korea
関連論文
- Efficient Implementation of Voiced/Unvoiced Sounds Classification Based on GMM for SMV Codec
- Forward Link Performance of Combined Soft and Hard Handoff in Multimedia CDMA Systems
- Output Characteristics of P-doped Raman Fiber Laser at 1484 nm with 2.11 W Maximum Output Power Pumped by CW 1064 nm Yb-doped Double-Clad Fiber Laser
- Output Characteristics of High-Power Continuous Wave Raman Fiber Laser at 1484 nm Using Phosphosilicate Fiber
- Output Characteristics of High-Power Continuous Wave Raman Fiber Laser at 1239 nm Using Phosphosilicate Fiber
- Simultaneous Double-Color Continuous Wave Raman Fiber Laser at 1239 nm and 1484 nm Using Phosphosilicate Fiber
- Ultra-Broadband CW Supercontinuum Generation Centered at 1483.4 nm from Brillouin/Raman Fiber Laser
- Flash-Lamp-Pumped Acousto-Optic Q-switched Cr-Tm:YAG Laser
- Theoretical and Experimental Investigations of Diode-Pumped Tm:YAG Laser in Active Mirror Configuration
- Her2^ genotype and breast cancer progression in Korean women
- Stability Limits of Swiss-roll Micro Combustors
- Emission and Efficiency of Small Swiss-Roll Combustors as Heat Sources
- Improved Global Soft Decision Using Smoothed Global Likelihood Ratio for Speech Enhancement(Multimedia Systems for Communications)
- Filtering of a Dissonant Frequency Based on Improved Fundamental Frequency Estimation for Speech Enhancement(Digital Signal Processing)
- A Statistical Model-Based V/UV Decision under Background Noise Environments(Speech and Hearing)
- A New Post-Filtering Algorithm for Residual Acoustic Echo Cancellation in Hands-Free Mobile Application(Mobile Multimedia Communications)
- Speech Reinforcement Based on Soft Decision under Far-End Noise Environments
- A Support Vector Machine-Based Gender Identification Using Speech Signal
- Frame Splitting Scheme for Error-Robust Audio Streaming over Packet-Switching Networks
- Integrated Dissonant Frequency Filtering and Noise Reduction for Improving Perceptual Quality of Noisy Speech and Husky Voice(Speech and Hearing)
- Acoustic Environment Classification Based on SMV Speech Codec Parameters for Context-Aware Mobile Phone
- Distorted Speech Rejection for Automatic Speech Recognition in Wireless Communication(Speech and Hearing)
- Indium Tin Oxide-Bus Separated Electrode Structure in Plasma Display Panel
- Improvement of SVM-Based Speech/Music Classification Using Adaptive Kernel Technique
- Speech Enhancement Based on Data-Driven Residual Gain Estimation
- Speech Enhancement Based on Adaptive Noise Power Estimation Using Spectral Difference
- A Statistical Model-Based Speech Enhancement Using Acoustic Noise Classification for Robust Speech Communication