http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 주제어 : Speaker verification system (검색결과 7 건)
- 무료
- 기관 내 무료
- 유료
-
위진우(Jinwoo Wee),김형배(Hyoung-Bae Kim) 산업기술교육훈련학회 2010 산업기술연구논문지 (JITR) Vol.15 No.3
When the design and implementation of speaker verification system with stable performances by short training number, the proper speaker recognition method of reducing the training number are required. For determining the proper speaker recognition method, the existing speaker recognition methods are compared through computer simulations. Discrete HMM recognition method is adopted for the performance stability of speaker recognition, the amount of training and the possibility of real-time implementation, The realistic methods are proposed for the enhancement of discrete HMM as the following five parts. Five algorithms are proposed in two parts and incorporated into a single system. This study consists of two parts; one is preprocessing part, the other is speaker verification part. The five methods mentioned above are applied to the general DHMM and the approach then results in the performance enhancement of about 1.1% in FA and 41.7% in FR. As a result, it is shown that the proposed speaker verification system composed of the five algorithms is very robust to noisy environment
-
김경화,소병민,유하진 한국음성학회 2012 말소리와 음성과학 Vol.4 No.3
In this paper, we introduce the automatic speaker identification system ‘SPO(Supreme Prosecutors Office) Verifier’. SPO Verifier is a GMM(Gaussian mixture model)-UBM(universal background model) based automatic speaker recognition system and has been developed using Korean speakers’ utterances. This system uses a channel compensation algorithm to compensate recording device characteristics. The system can give the users the ability to manage reference models with utterances from various environments to get more accurate recognition results. To evaluate the performance of SPO Verifier on Korean speakers, we compared this system with one of the most widely used commercial systems in the forensic field. The results showed that SPO Verifier shows lower EER(equal error rate) than that of the commercial system.
-
Voice Playback Detection based on Long-window Scale-factors
U. Nanaji,N. Thirupathi Rao,Ch. Raj Kumar,Debnath Bhattacharyya,Hye-jin Kim 보안공학연구지원센터 2016 International Journal of Security and Its Applicat Vol.10 No.12
With the popularity of high fidelity and portable recording device, it becomes very easy for the attacker to spoof speaker verification system by voice playback. In this paper, it found that there exists obvious difference in the scale-factors, a parameter of MP3 codec, when the original and playback voices are compressed by MP3. So, a detection algorithm for the playback voice is proposed. The experimental results show that the detection accuracy of the proposed algorithm can reach to 99.51% for 4 different types of eavesdropping devices. Meanwhile, the algorithm is integrated into the speaker verification system based on GMM-UBM. The equal error rate (EER) of the system has dropped 32% and the system's ability to resist playback attack is improved.
-
DSP보드를 이용한 전화음성용 실시간 화자인증 시스템의 구현에 관한 연구
이현승,최홍섭,Lee Hyeon Seung,Choi Hong Sub 대한음성학회 2004 말소리 Vol.49 No.-
This paper is aiming at implementation of real-time speaker verification system using DSP board. Dialog/4, which is based on microprocessor and DSP processor, is selected to easily control telephone signals and to process audio/voice signals. Speaker verification system performs signal processing and feature extraction after receiving voice and its ID. Then through computing the likelihood ratio of claimed speaker model to the background model, it makes real-time decision on acceptance or rejection. For the verification experiments, total 15 speaker models and 6 background models are adopted. The experimental results show that verification accuracy rates are 99.5% for using telephone speech-based speaker models.
-
On-Line Linear Combination of Classifiers Based on Incremental Information in Speaker Verification
Fernando Huenupan,Néstor Becerra Yoma,Carlos Molina,Claudio Garretón 한국전자통신연구원 2010 ETRI Journal Vol.32 No.3
A novel multiclassifier system (MCS) strategy is proposed and applied to a text-dependent speaker verification task. The presented scheme optimizes the linear combination of classifiers on an on-line basis. In contrast to ordinary MCS approaches, neither a priori distributions nor pre-tuned parameters are required. The idea is to improve the most accurate classifier by making use of the incremental information provided by the second classifier. The on-line multiclassifier optimization approach is applicable to any pattern recognition problem. The proposed method needs neither a priori distributions nor pre-estimated weights, and does not make use of any consideration about training/testing matching conditions. Results with Yoho database show that the presented approach can lead to reductions in equal error rate as high as 28%, when compared with the most accurate classifier, and 11% against a standard method for the optimization of linear combination of classifiers.
-
최형기 ( Hyung-ki Choi ),박기영 ( Kee-young Park ) 한국고등직업교육학회 2002 한국고등직업교육학회논문집 Vol.3 No.2
This paper has performed the speech security system which is operated by software program using DTW based text prompt speaker verification algorithm in accordance with 12th gammatone filter frequency cepstrum coefficients for feature parameter. This realized system has operated sequentially as below: Users utter the proposed words, the uttered word is transformed as the feature parameters, the pattern is compared with the already stored reference pattern and could decide the user's availability whether use it or not. This research executed 10 words from 15 speakers as test data for the security system, experimentally. The result showed FA 10.2% and FR 8.84% respectively, which are average errors in the simulated experimental circumstances.
-
지진구,육상조,윤성일 공주영상정보대학 1998 논문집 Vol.5 No.-
본 논문에서는 자연스럽고 사용하기 쉬운 각 개인의 음성을 이용하여 신원을 확인 할 수 있는 화자 검증 시스템을 구현하였다. 본 연구에서 구현한 시스템은 상대적인 대수 에너지만을 사용하여 개략적인 끝점 검출을 실행하였고, 검출된 음성부분에 대하여 몇 가지 특징 파라메터에 대한 검증율 실험의 결과를 바탕으로 특징 파라메터로는 필터 뱅크 계수를 사용하였다. 또한 자신이 등록한 음성 패스워드로 각각 개인의 코드북을 따로 구성하는 방법을 사용하여 화자 검증율을 높였다. 벡터양자화에 대응하여 동작하는 상부 인식기로는 소규모 고립단어 인식에 적합한 동적 시간 정합법을 채택하였다. 본 논문에서는 효과적인 화자 검증 시스템을 구현하기 위한 특징 파라메터 및 코드북 구성에 따른 상대적인 검증율을 실험을 통하여 측정하여, 음성을 등록한 사람에 대해 각각의 코드북을 생성하여 화자 검증 시스템을 구현하였다. 향후 연구과제로 본 시스템은 잡음에 대해서 고려하지 않았지 때문에 잡음 환경에서도 사용자의 실제 음성을 이용하여 화자 검증 할 수 있도록 개선이 있어야 할 것이며, 전자상거래 및 정보보안에서의 전자 서명과 같이 음성을 이용하여 신원을 확인하는 기타 응용분야에 대한 연구가 필요하다.
ScienceON연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
