Tolerance Interval Analysis를 이용한 배경화자 없는 간단한 화자인증시스템에 관한 연구

최홍섭,Choi, Hong-Sub 대한음성학회 2005 말소리 Vol.56 No.-

In this paper, we are focused to develop the simplified speaker verification algorithm without background speaker models, which will be adopted in the portable speaker verification system equipped in portable terminals such as mobile phone and PMP. According to the tolerance interval analysis, the population of someone's speaker model can be represented by a suitable number of selected independent samples of speaker model. So we can make the representative speaker model and threshold under the specified confidence level and coverage. Using proposed algorithm with the number of samples is 40, the experiments show that the false rejection rate is $3.0\%$ and the false acceptance rate $4.3\%$, worth comparing to conventional method's results, $5.4\%\;and\;5.5\%$, respectively. Next step of research will be on the suitable adaptation methods to overcome speech variation problems due to aging effect and operating environments.

휴대폰음성을 이용한 화자인증시스템에서 배경화자에 따른 성능변화에 관한 연구

최홍섭,Choi, Hong-Sub 한국음성학회 2005 음성과학 Vol.12 No.3

It was verified that a speaker verification system improved its performances of EER by regularizing log likelihood ratio, using background speaker models. Recently the wireless mobile phones are becoming more dominant communication terminals than wired phones. So the need for building a speaker verification system on mobile phone is increasing abruptly. Therefore in this paper, we had some experiments to examine the performance of speaker verification based on mobile phone's voices. Especially we are focused on the performance variations in EER(Equal Error Rate) according to several background speaker's characteristics, such as selecting methods(MSC, MIX), number of background speakers, aging factor of speech database. For this, we constructed a speaker verification system that uses GMM(Gaussin Mixture Model) and found that the MIX method is generally superior to another method by about 1.0% EER. In aspect of number of background speakers, EER is decreasing in proportion to the background speakers populations. As the number is increasing as 6, 10 and 16, the EERs are recorded as 13.0%, 12.2%, and 11.6%. An unexpected results are happened in aging effects of the speech database on the performance. EERs are measured as 4%, 12% and 19% for each seasonally recorded databases from session 1 to session 3, respectively, where duration gap between sessions is set by 3 months. Although seasons speech database has 10 speakers and 10 sentences per each, which gives less statistical confidence to results, we confirmed that enrolled speaker models in speaker verification system should be regularly updated using the ongoing claimant's utterances.

SNR을 이용한 프레임별 유사도 가중방법을 적용한 문맥종속 화자인식에 관한 연구

최홍섭,Choi, Hong-Sub 대한음성학회 2007 말소리 Vol.61 No.-

The environmental differences between training and testing mode are generally considered to be the critical factor for the performance degradation in speaker recognition systems. Especially, general speaker recognition systems try to get as clean speech as possible to train the speaker model, but it's not true in real testing phase due to environmental and channel noise. So in this paper, the new method of weighting the frame-based likelihood according to frame SNR is proposed in order to cope with that problem. That is to make use of the deep correlation between speech SNR and speaker discrimination rate. To verify the usefulness of this proposed method, it is applied to the context dependent speaker identification system. And the experimental results with the cellular phone speech DB which is designed by ETRI for Koran speaker recognition show that the proposed method is effective and increase the identification accuracy by 11% at maximum.

WHMM에 적용가능한 MCE/GPD 학습알고리듬에 관한 연구

최홍섭,Choi, Hong-Sub 한국음향학회 1997 韓國音響學會誌 Vol.16 No.1

This paper derives a new training alorithm for WHMM using the well-known MCE/GPD method with experimental results on the E-set. The derived algorithm generalizes the conventional adaptive training algorithm for WHMM, which means that HMMs of multiple competing classes can be trained at the same time. The recognition results on the E-set have shown about 15% and 12% improvement for training and test data, respectively. 본 논문에서는 잘 알려진 변별학습 방법인 MCE/GPD방법을 WHMM에 적용시켜 그 학습알고리듬을 유도하고, E-set에 대한 실험결과를 제시한다. 유도된 알고리듬은 여러개의 혼동 가능한 클래스의 HMM에 대해서 학습이 가능하다는 점에서 기존에 제시된 적응학습 알고리듬의 자연스러운 일반화라 할 수 있다. E-set에 대한 인식실험 결과 학습에이터에 대해서 15%, 시험데이터에 대해서 12% 정도의 인식율 개선을 얻을 수 있었다.

음성천이구간에서의 성도 파라메타 시변추정에 관한 연구

최홍섭,Choi, Hong-Sub 한국음향학회 1997 韓國音響學會誌 Vol.16 No.2

In this paper, sample selective RLS(SSRLS) method is proposed, which aims to eliminate the influence of pitch bias. Its basic concepts are as follows. First it extracts the open glottis interval by using the residual signals, then estimates the formant values from the selected speech samples excluding above open glottis interval. This method has some analogy with the SSLPS, the simulation is conducted upon the synthetic and real speech. From these results, we find more usefulness of the proposed method than the conventional ones. 음성의 천이구간에서의 특징 파라메타를 찾아내기 위하여 본 논문에서는 AR모델을 사용하여 적응적으로 성문폐쇄구간을 찾은 후, 이를 제외한 구간에서 성도 파라메타를 추정함으로써 음원의 피치바이어스 영향을 제거하는 SSRLS(Sample Selective RLS)방법을 제안한다. 성능을 비교하기 위하여 합성음과 실제음에 대하여 포만트 추정실험을 했으며, 실험결과 제안된 방법이 WRLS 보다 우수함을 알 수 있었다.

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.

2차원 상의 음원위치 추정을 위한 효율적인 영역분할방법

김환용,최홍섭,Kim, Hwan-Yong,Choi, Hong-Sub 한국음향학회 2016 韓國音響學會誌 Vol.35 No.5

음원의 위치를 찾는 SSL(Sound Source Localization)은 로봇과의 인터페이스, 화상회의, 스마트 자동차 등 여러 분야에서 꼭 필요한 기술이다. 일반적으로 음원의 위치 정보를 활용하는 기술들은 주로 측정 장치에 대한 음원의 각도 정보를 찾아서 이용하고 있다. 그러나 음원의 위치에 대한 각도를 추정할 때 이용하는 사인 역함수의 비선형적인 특성으로 추정된 각도에 오차가 발생하며, 이에 대한 방안으로 마이크가 담당하는 영역을 분할하는 방법이 제안되었다. 본 논문에서는 마이크 어레이 패턴에 따른 영역분할 방법을 제안하고 음원의 위치를 2차원상의 평면 좌표로 특정하는 방법으로 위치 추정 성능을 평가하였다. 실험에서 잡음에 강인한 GCC-PHAT(Generalized Cross Correlation Phase Transform) 방법을 사용했으며, 마이크 어레이의 패턴은 마이크 3개와 4개로 삼각형과 사각형 두 종류로 구성하였으며, 100개의 음성 데이터로 실험한 결과 실제 환경에서는 3개의 마이크 어레이를 사용해서는 영역 분할 해상도가 낮아서 음원의 위치를 정해진 특정 범위내로 추정하는데 실패했으나, 4개 마이크를 이용하여 해상도를 높였더니 위치추정 성공률이 67 %로 크게 향상됨을 확인할 수 있었다. SSL (Sound Source Localization) has been applied to several applications such as man-machine interface, video conference system, smart car and so on. But in the process of sound source localization, angle estimation error is occurred mainly due to the non-linear characteristics of the sine inverse function. So an approach was proposed to decrease the effect of this non-linear characteristics, which divides the microphone's covering space into narrow regions. In this paper, we proposed an optimal space dividing way according to the pattern of microphone array. In addition, sound source's 2-dimensional position is estimated in order to evaluate the performance of this dividing method. In the experiment, GCC-PHAT (Generalized Cross Correlation PHAse Transform) method that is known to be robust with noisy environments is adopted and triangular pattern of 3 microphones and rectangular pattern of 4 microphones are tested with 100 speech data respectively. The experimental results show that triangular pattern can't estimate the correct position due to the lower space area resolution, but performance of rectangular pattern is dramatically improved with correct estimation rate of 67 %.

DSP 프로세서를 이용한 실시간 ANC 시스템 구현에 관한 연구

이영일,최홍섭,Lee Young Il,Choi Hong Sub 대한음성학회 2004 말소리 Vol.52 No.-

This paper is aiming at real-time implementation of adaptive noise cancellation system using DSP processor. ACHARF algorithm, which guarantees stability and fast convergence by adaptive compensator, is used on this DSP system. For the experiments, TLV320AIC23 stereo CODEC of TI Inc. is used with TMS320C6413 DSP processor. Signals of primary input and reference input are obtained by two microphones. The primary input is the voice plus noise signal and the reference input is white noise or real noise. The experimental results show that ANC system using DSP processor with ACHARF is verified to be an effective speech enhancement method for various speech processing units.

자동차 환경에서 TDOA를 이용한 화자위치추정 방법

이상헌(Sang-Hun Lee),최홍섭(Hong-Sub Choi) 한국디지털콘텐츠학회 2016 한국디지털콘텐츠학회논문지 Vol.17 No.2

휴대용 멀티미디어 디바이스를 위한 TPO(Time, Place, Occasion)-Shift 시스템 설계에 대한 연구

김대진(Dae-Jin Kim),최홍섭(Hong-Sub Choi) 한국컴퓨터정보학회 2009 韓國컴퓨터情報學會論文誌 Vol.14 No.2

광대역 네트워크의 발달과 함께 멀티미디어 산업의 발달은 IPTV와 같은 디지털 콘텐츠 시장의 확산을 가져오고 있다. 이러한 배경 속에서 멀티미디어에 대한 욕구를 만족시키기 위해 Time-Shift 시스템이 개발되었다. 이 시스템은 시간(Time)에 대한 독립적 특성만 강조되었기 때문에 장소(Place)와 상황(Occasion)에 대해서는 독립적이지 못하다. 이러한 문제를 해결하기 위해서 본 논문에서는 휴대용 멀티미디어 디바이스를 위한 TPO(Time, Place, Occasion)-Shift 시스템에 대한 설계를 제안한다. 휴대용 멀티미디어 디바이스에서 지원 가능한 프로파일과 일반 셋탑박스에서 지원 가능한 프로파일은 다르고, 휴대용 디바이스에서는 대용량의 멀티미디어 데이터를 무한정으로 저장할 수 없기 때문에 한정된 양의 데이터를 원하는 프로파일로 계속저장하는 것이 설계의 핵심이다. 따라서 보다 효율적인 버퍼 관리를 위해서 지정시간 단위의 바스켓을 구성하고, 바스켓의 파일이름에 시간정보를 삽입함으로써 새로운 콘텐츠 구성을 위한 DTS(Decoding Time Stamp) 정보로 파일이름을 사용할 수 있다. 따라서 트랜스코딩을 통하여 데이터 변환할 때, DTS정보를 이용하여 새로운 포맷의 콘텐츠를 휴대용 멀티미디어 디바이스에 구성할 수 있게 된다. 또한 바스켓 기반의 버퍼시스템를 이용하여 모바일 디바이스에 실시간으로 컨텐츠를 구성하고 셋탑박스내에서 메모리를 적게 사용한다. 본 논문에서 제안한 TPO-Shift시스템은 윈도우즈 비스타의 환경에서 다이렉트쇼(Directshow) 재생기를 이용한 셋탑박스, 그리고 휴대용 디바이스인 MS340 단말기로 구현하였으며, 실시간으로 TPO-Shift 시스템을 충분히 실행함을 확인할 수 있었다. While the broadband network and multimedia technology are being developed, the commercial market of digital contents as well as using IPTV has been widely spreading. In this background, Time-Shift system is developed for requirement of multimedia. This system is independent of Time but is not independent of Place and Occasion. For solving these problems, in this paper, we propose the TPO(Time, Place, Occasion)-Shift system for mobile multimedia devices. The profile that can be applied to the mobile multimedia devices is much different from that of the setter-box. And general mobile multimedia devices could not have such large memories that is for multimedia data. So it is important to continuously store and manage those multimedia data in limited capacity with mobile device's profile. Therefore we compose the basket in a way using defined time unit and manage these baskets for effective buffer management. In addition. since the file name of basket is made up to include a basket's time information, we can make use of this time information as DTS(Decoding Time Stamp). When some multimedia content is converted to be available for portable multimedia devices, we are able to compose new formatted contents using such DTS information. Using basket based buffer systems, we can compose the contents by real time in mobile multimedia devices and save some memory. In order to see the system's real-time operation and performance, we implemented the proposed TPO-Shift system on the basis of mobile device, MS340. And setter-box are desisted by using directshow player under Windows Vista environment. As a result, we can find the usefulness and real-time operation of the proposed systems.