Efficient Speech Reinforcement Based on Low-Bit-Rate Speech Coding Parameters
スポンサーリンク
概要
- 論文の詳細を見る
In this paper, a novel approach to speech reinforcement in a low-bit-rate speech coder under ambient noise environments is proposed. The excitation vector of ambient noise is efficiently obtained at the near-end and then combined with the excitation signal of the far-end for a suitable reinforcement gain within the G.729 CS-ACELP Annex. B framework. For this reason, this can be clearly different from previous approaches in that the present approach does not require an additional arithmetic step such as the discrete Fourier transform (DFT). Experimental results indicate that the proposed method shows better performance than or at least comparable to conventional approaches with a lower computational burden.
- (社)電子情報通信学会の論文
- 2010-09-01
著者
-
CHANG Joon-Hyuk
Department of Electronic Engineering, Inha University
-
Chang Joon-hyuk
Department Of Electrical And Computer Engineering University Of California At Santa Barbara
-
CHOI Jae-Hun
Department of Electronic Engineering, Inha University
-
LEE Seong-Ro
Mokpo National University
-
Choi Jae-hun
Department Of Electronic Engineering Inha University
関連論文
- Discriminative Weight Training for Support Vector Machine-Based Speech/Music Classification in 3GPP2 SMV Codec
- Discriminative Weight Training for Support Vector Machine-Based Speech/Music Classification in 3GPP2 SMV Codec
- Speech/Music Classification Enhancement for 3GPP2 SMV Codec Based on Support Vector Machine
- A Statistical Model-Based V/UV Decision under Background Noise Environments(Speech and Hearing)
- Efficient Speech Reinforcement Based on Low-Bit-Rate Speech Coding Parameters
- Multiband Vector Quantization Based on Inner Product for Wideband Speech Coding(Speech and Hearing)
- A Novel Approach to a Robust a Priori SNR Estimator in Speech Enhancement(Multimedia Systems for Communications)
- A Support Vector Machine-Based Voice Activity Detection Employing Effective Feature Vectors
- Improved Global Soft Decision Incorporating Second-Order Conditional MAP in Speech Enhancement