• The Journal of the Acoustical Society of Korea
• /
• v.23 no.1
• /
• pp.61-74
• /
• 2004

In this paper, a contort- and speaker-dependent cepstrum extraction method and a channel normalization method for minimizing the loss of speaker characteristics in the cepstrum were proposed for a robust speaker recognition system over the channel. The proposed extraction method creates a cepstrum based on the pitch synchronous analysis using the inherent pitch of the speaker. Therefore, the cepstrum called the 〃pitch synchronous cepstrum〃 (PSC) represents the impulse response of the vocal tract more accurately in voiced speech. And the PSC can compensate for channel distortion because the pitch is more robust in a channel environment than the spectrum of speech. And the proposed channel normalization method, the 〃formant-broadened pitch synchronous CMS〃 (FBPSCMS), applies the Formant-Broadened CMS to the PSC and improves the accuracy of the intraframe processing. We compared the text-independent closed-set speaker identification on 56 females and 112 males using TIMIT and NTIMIT database, respectively. The results show that pitch synchronous km improves the error reduction rate by up to 7.7% in comparison with conventional short-time cepstrum and the error rates of the FBPSCMS are more stable and lower than those of pole-filtered CMS.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.39-42
• /
• 2000

A bootstrap and aggregating (bagging) vector quantization (VQ) classifier is proposed for speaker recognition. This method obtains multiple training data sets by resampling the original training data set, and then integrates the corresponding multiple classifiers into a single classifier. Experiments involving a closed set, text-independent and speaker identification system are carried out using the TIMIT database. The proposed bagging VQ classifier shows considerably improved performance over the conventional VQ classifier.

• Journal of the Korea Computer Industry Society
• /
• v.2 no.10
• /
• pp.1245-1252
• /
• 2001

In this paper, we have identified speakers who give digits in Korean. In order to identify speakers, we have utilized the specifie feature parameters which extracted from sound wave. We have noticed that multipulses are present in pitch periods of sound wave, which containes the personal information and depends on the speakers. In this experiment, we have extracted multipulses, and have attempted to identify the speaker by investigating the specific feature parameters of each speaker based on the extracted multipulses.

• Speech Sciences
• /
• v.7 no.2
• /
• pp.127-137
• /
• 2000

This paper presents a macro script to analyze numeric sounds by a speech analysis shareware, Praat, and analyzes those sounds produced by three students who were born and raised in Pusan. Recording was done in a quiet office. To make a meaningful comparison, dynamic time points in relation to the total duration of voicing segments were determined to measure acoustical values. Results showed that a strong correlation coefficient was found between the repetitive production of numeric sounds within and across the speakers. Very high coefficients among diphthongal numbers (0 and 6) which usually show wide formant variation were noticed. This supports that each speaker produced numbers quite coherently. Also, the frequency differences between the three subjects were within a perceptually similar range. To identify a speaker among others may require to find subtle individual differences within this range. Perceptual experiments by synthesized numeric sounds may lead to resolve the issue.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2447-2450
• /
• 2003

화자인식은 음성의 특성을 이용해서 화자의 신원을 확인하는 기술이다. 이러한 기술은 등록된 화자집단 중 화자를 식별하는 화자식별(speaker Identification)과 지금 발성한 화자만을 비교하여 확인하는 화자확인(speaker verification)이 있다. 이러한 화자인식은 음성에 내재되어 있는 화자정보를 추출하여 개인을 확인하는 기술로 전화망을 통한 서비스가 확산되어 가고 있는 현대사회에 가장 효과적인 기술 중 하나이다. 또한 PDA를 이용한 증건거래 시스템 등 현대사회에서는 실시간으로 화자인식이 이루어져야 한다. 본 논문에서는 이와 같이 실시간 화자인식을 위한 처리시간 단축에 관하여 연구하였다. 처리시간 단축을 위하여 우선 피치주기 단위로 음성 파형을 분해한 다음 분해된 피치 단위에 윈도우 함수를 곱해서 단구간 신호의 열로 만들고 분해된 단위를 조절하는 PSOLA 합성방식을 이용하여 인식 시스템의 전처리단을 재구성하였다. 이와 같은 방식으로 제안한 인식시스템의 처리시간, 인식률을 기존의 화자인식 시스템과 비교하였다.

• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.393-394
• /
• 2007

Speaker Recognition for the Intelligent Service Robot is implemented in this paper. For this purpose, we perform speaker recognition based on Gaussian Mixture Model(GMM) and use robot platform called WEVER, which is a Ubiquitous Robotic Companion(URC) intelligent service robot developed at Intelligent Robot Research Division in ETRI. The experimental results reveals that the approach presented in this paper yields a good identification (89.00%) performance within 2 meter distance.

• Speech Sciences
• /
• v.8 no.3
• /
• pp.43-50
• /
• 2001

This paper deals with voice similarities between sisters who are supposed to have common physiological characteristics from a single biological mother. Nine pairs of sisters who are believed to have similar voices participated in this experiment. The speech samples obtained from one pair of sisters were eliminated in the analysis because their perceptual score was relatively low. The words were measured in both isolation and context, and the subjects were asked to read the text five times with about three seconds of interval between readings. Recordings were made at natural speed in a quiet room. The data were analyzed in pitch and formant frequencies using CSL (Computerized Speech Lab) and PCQuirer. It was found that data of the initial vowels are much more similar and homogeneous than those of vowels in other positions. The acoustic data showed that voice similarities are strikingly high in both pitch and formant frequencies. It is assumed that statistical data obtained from this experiment can be used as a guideline for modelling speaker identification and speaker verification.

• ETRI Journal
• /
• v.28 no.3
• /
• pp.320-328
• /
• 2006

Using biometrics to verify a person's identity has several advantages over the present practice of personal identification numbers (PINs) and passwords. To gain maximum security in a verification system using biometrics, the computation of the verification as well as the storing of the biometric pattern has to take place in a smart card. However, there is an open issue of integrating biometrics into a smart card because of its limited resources (processing power and memory space). In this paper, we propose a speaker verification algorithm using a support vector machine (SVM) with a very few features, and implemented it on a 32-bit smart card. The proposed algorithm can reduce the required memory space by a factor of more than 100 and can be executed in real-time. Also, we propose a hardware design for the algorithm on a field-programmable gate array (FPGA)-based platform. Based on the experimental results, our SVM solution can provide superior performance over typical speaker verification solutions. Furthermore, our FPGA-based solution can achieve a speed-up of 50 times over a software-based solution.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.2
• /
• pp.184-189
• /
• 2013

In this paper, we propose the injurious moving picture filtering system using certain sounds contained in the injurious moving picture to filter injurious moving picture which is distributed without limitation in internet and internet storage space. For this purpose, the Gaussian Mixture Model which can well represent the characteristics of the sound, is used and frame level likelihood estimation is used to calculate the likelihood between filtering target data and the sound models. Also, the pruning method which can real-time proceed by reducing the comparing number of data, is applied for real-time processing, and MWMR method which showed good performance from existing speaker identification, is applied for the distinguish performance of high precision. In the identification experiment result, in case of the frame rate which is the proportion of total frame to high likelihood frame, is set to 50%, identification error rate is 6.06%, and in case of frame rate is set to 60%, error rate is 3.03%. As the result, the proposed system can distinguish between general and injurious moving picture effectively.

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.1
• /
• pp.8-14
• /
• 2002

In this paper, we realized a real-time text-independent speaker recognition system using gaussian mixture model, and applied frame level likelihood normalization method which shows its effects in verification system. The system has three parts as front-end, training, recognition. In front-end part, cepstral mean normalization and silence removal method were applied to consider speaker's speaking variations. In training, gaussian mixture model was used for speaker's acoustic feature modeling, and maximum likelihood estimation was used for GMM parameter optimization. In recognition, likelihood score was calculated with speaker models and test data at frame level. As test sentences, we used text-independent sentences. ETRI 445 and KLE 452 database were used for training and test, and cepstrum coefficient and regressive coefficient were used as feature parameters. The experiment results show that the frame-level likelihood method's recognition result is higher than conventional method's, independently the number of registered speakers.