• MALSORI
• /
• no.57
• /
• pp.175-189
• /
• 2006

Because a packet loss brings about degradation of speech quality, VoIP speech coders have PLC (Packet Loss Concealment) mechanism. G.711, which is a mandatory VoIP speech coder, also has the PLC algorithm based on pitch period replication. However, it is not robust to burst losses. Thus, we propose two methods to improve the performance of the original PLC algorithm in G.711. One adaptively utilizes voiced/unvoiced information of adjacent good frames regarding to the current lost frame. The other is based on adaptive gain control according to energy variation across the frames. We evaluate the performance of the proposed PLC algorithm by measuring a PESQ value under different random and burst packet loss simulating conditions. It is shown from the experiments that the performance of the proposed PLC algorithm outperforms that of PLC employed in ITU-T Recommendation G.711.

• Proceedings of the KSPS conference
• /
• 2007.05a
• /
• pp.307-310
• /
• 2007

In this paper, a packet loss concealment (PLC) algorithm for CELP-type speech coders is proposed to improve the quality of decoded speech under a burst packet loss condition. The proposed algorithm is based on the recovery of voiced excitation using an estimate of the voicing probability and the generation of random excitation by permutating the previously decoded excitation. The voicing probability is estimated from the correlation using the previous correctly decoded excitation and pitch. The proposed algorithm is implemented as a PLC algorithm for G.729 and its performance is compared with PLC employed in G.729 by means of perceptual evaluation of speech quality (PESQ) and an A-B preference test under the random and burst packet losses with rates of 3% and 5%. It is shown that the proposed algorithm provides better speech quality than the PLC of G.729, especially under burst pack losses.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.10
• /
• pp.4930-4951
• /
• 2018

The deep learning based speech enhancement has shown considerable success. However, it still suffers performance degradation under mismatch conditions. In this paper, an adaptation method is proposed to improve the performance under noise mismatch conditions. Firstly, we advise a noise aware training by supplying identity vectors (i-vectors) as parallel input features to adapt deep neural network (DNN) acoustic models with the target noise. Secondly, given a small amount of adaptation data, the noise-dependent DNN is obtained by using $L_2$ regularization from a noise-independent DNN, and forcing the estimated masks to be close to the unadapted condition. Finally, experiments were carried out on different noise and SNR conditions, and the proposed method has achieved significantly 0.1%-9.6% benefits of STOI, and provided consistent improvement in PESQ and segSNR against the baseline systems.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.2E
• /
• pp.35-42
• /
• 2009

In this paper, we present criteria of objective measurement of speech quality to provide the mobile-VoIP services efficiently over wireless mobile internet. The mobile-VoIP service, which is based on mobility and is error-prone compared to conventional VoIP over wired network, is about to be launched, but there have not been adequate quality indexes and the Quality of Service (QoS) standards for evaluating speech quality of Mobile-VoIP. In addition, there are many factors influencing on the speech quality in packet network of which packet loss contribute directly to the overall voice communication quality. For this reason, we adopt the Gilbert-Elliot Channel Model for modeling packet network based on IP and assess the voice quality through the objective speech method of ITU-T P. 862 PESQ and ITU-T P. 862.1 MOS-LQO under various packet loss rates in the transmission channel environments. Our simulation results address the specific criteria and QoS for the mobile-VoIP services in terms of the various packet loss environments.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.12
• /
• pp.1920-1925
• /
• 2012

In this paper, we propose a novel method to improve the performance of the improved minima controlled recursive averaging (IMCRA) in non-stationary noisy environment. The conventional IMCRA algorithm efficiently estimate the noise power by averaging past spectral power values based on a smoothing parameter that is adjusted by the signal presence probability in frequency subbands. Since the minimum of smoothing parameter is defined as 0.85, it is difficult to obtain the robust estimates of the noise power in non-stationary noisy environments that is rapidly changed the spectral characteristics such as babble noise. For this reason, we proposed the modified IMCRA, which adaptively estimate and updata the noise power according to the noise type classified by the Gaussian mixture model (GMM). The performances of the proposed method are evaluated by perceptual evaluation of speech quality (PESQ) and composite measure under various environments and better results compared with the conventional method are obtained.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.07a
• /
• pp.418-420
• /
• 2015

본 논문은 베이지안 비음수 행렬 인수분해 (Bayesian nonnegative matrix factorization, BNMF) 기반의 음성 강화 기법에서 음성과 잡음 성분의 latent source 수에 따른 강화성능에 대해 서술한다. BNMF 기반의 음성 강화 기법은 입력 신호를 서브 신호들의 합으로 분해한 후, 잡음 성분을 제거하는 방식으로 그 성능이 기존의 NMF 기반의 방법들보다 우수한 것으로 알려져 있다. 그러나 많은 계산량과 latent source 의 수에 따라 성능의 차이가 있다는 단점이 있다. 이러한 단점을 개선하기 위해 본 논문에서는 BNMF 기반의 음성 강화 기법에서 최적의 latent source 개수를 찾기 위한 실험을 진행하였다. 실험은 잡음의 종류, 음성의 종류, 음성과 잡음의 latent source 의 개수, 그리고 SNR 을 바꿔가며 진행하였고, 성능 평가 방법으로 PESQ (perceptual evaluation of speech quality) 를 이용하였다. 실험 결과, 음성의 latent source 개수는 성능에 영향을 주지 않지만, 잡음의 latent source 개수는 많을수록 성능이 좋은 것으로 확인되었다.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.12
• /
• pp.1756-1760
• /
• 2015

This paper focuses on a priori signal to noise ratio (SNR) estimation method for the speech enhancement. There are many researches for speech enhancement with several ambient noise cancellation methods. The method based on spectral subtraction (SS) which is widely used in noise reduction has a trade-off between the performance and the distortion of the signals. So the need of adaptive method like an estimated a priori SNR being able to making a high performance and low distortion is increasing. The decision directed (DD) approach is used to determine a priori SNR in noisy speech signals. A priori SNR is estimated by using only the magnitude components and consequently follows a posteriori SNR with one frame delay. We propose a modified a priori SNR estimator and the weighted rational transfer function for speech enhancement with wind noises. The experimental result shows the performance of our proposed estimator is better Perceptual Evaluation of Speech Quality scores (PESQ, ITU-T P.862) compare to the conventional DD approach-based systems and different noise reduction methods.

• MALSORI
• /
• no.54
• /
• pp.27-43
• /
• 2005

Existing standard speech coders can provide speech communication of high quality while they degrade the performance of speech recognition systems that use the reconstructed speech by the coders. The main cause of the degradation is that the spectral envelope parameters in speech coding are optimized to speech quality rather than to the performance of speech recognition. For example, mel-frequency cepstral coefficient (MFCC) is generally known to provide better speech recognition performance than linear prediction coefficient (LPC) that is a typical parameter set in speech coding. In this paper, we propose a speech coder using MFCC instead of LPC to improve the performance of a server-based speech recognition system in network environments. However, the main drawback of using MFCC is to develop the efficient MFCC quantization with a low-bit rate. First, we explore the interframe correlation of MFCCs, which results in the predictive quantization of MFCC. Second, a safety-net scheme is proposed to make the MFCC-based speech coder robust to channel error. As a result, we propose a 8.7 kbps MFCC-based CELP coder. It is shown from a PESQ test that the proposed speech coder has a comparable speech quality to 8 kbps G.729 while it is shown that the performance of speech recognition using the proposed speech coder is better than that using G.729.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.335-337
• /
• 2018

음성향상기법은 음성에 포함된 잡음이나 잔향을 제거하는 기술로써 마이크로폰으로 입력된 음성신호는 잡음이나 잔향에 의해 왜곡되어지므로 음성인식, 음성통신 등의 음성신호처리 기술의 핵심 기술이다. 이전에는 음성신호와 잡음신호 사이의 통계적 정보를 이용하는 통계모델 기반의 음성향상기법이 주로 사용되었으나 통계 모델 기반의 음성향상기술은 정상 잡음 환경과는 달리 비정상 잡음 환경에서 성능이 크게 저하되는 문제점을 가지고 있었다. 최근 머신러닝 기법인 심화신경망 (DNN, deep neural network)이 도입되어 음성 향상 기법에서 우수한 성능을 내고 있다. 심화신경망을 이용한 음성 향상 기법은 다수의 은닉 층과 은닉 노드들을 통하여 잡음이 존재하는 음성 신호와 잡음이 존재하지 않는 깨끗한 음성 신호 사이의 비선형적인 관계를 잘 모델링하였다. 이러한 심화신경망 기반의 음성향상기법을 향상 시킬 수 있는 방법 중 하나인 강화학습을 적용하여 기존 심화신경망 대비 성능을 향상시켰다. 강화학습이란 대표적으로 구글의 알파고에 적용된 기술로써 특정 state에서 최고의 reward를 받기 위해 어떠한 policy를 통한 action을 취해서 다음 state로 나아갈지를 매우 많은 경우에 대해 학습을 통해 최적의 action을 선택할 수 있도록 학습하는 방법을 말한다. 본 논문에서는 composite measure를 기반으로 reward를 설계하여 기존 PESQ (Perceptual Evaluation of Speech Quality) 기반의 reward를 설계한 기술 대비 음성인식 성능을 높였다.

• The Journal of the Acoustical Society of Korea
• /
• v.29 no.1
• /
• pp.82-90
• /
• 2010

This paper proposes an efficient method to enhance the quality of synthesized speech in HMM based speech synthesis system. The proposed method trains spectral parameters and excitation signals using Gaussian mixture model, and estimates appropriate excitation signals from spectral parameters during the synthesis stage. Both WB-PESQ and MUSHRA results show that the proposed method provides better speech quality than conventional HMM based speech synthesis system.