• 음성과학
• /
• 제14권2호
• /
• pp.105-114
• /
• 2007

The performance of speaker identifier is severely degraded in noisy environments. A study suggested the concept of observation membership for enhancing performances of speaker identifier with noisy speech [1]. The method scaled observation probabilities of input speech by observation identification values decided by SNR. In the paper [1], the authors suggested heuristic parameter values for membership function. In this paper we attempt to apply particle swarm optimization (PSO) for obtaining the optimal parameters for speaker identification in noisy environments. With the speaker identification experiments using the ETRI database we prove that the optimization approach can yield better performance than using only the original membership function.

• The Journal of the Acoustical Society of Korea
• /
• 제21권4E호
• /
• pp.156-163
• /
• 2002

We apply independent component analysis (ICA) for extracting an optimal basis to the problem of finding efficient features for representing speech signals of a given speaker The speech segments are assumed to be generated by a linear combination of the basis functions, thus the distribution of speech segments of a speaker is modeled by adapting the basis functions so that each source component is statistically independent. The learned basis functions are oriented and localized in both space and frequency, bearing a resemblance to Gabor wavelets. These features are speaker dependent characteristics and to assess their efficiency we performed speaker identification experiments and compared our results with the conventional Fourier-basis. Our results show that the proposed method is more efficient than the conventional Fourier-based features in that they can obtain a higher speaker identification rate.

• 한국음향학회지
• /
• 제21권4호
• /
• pp.156-156
• /
• 2002

We apply independent component analysis (ICA) for extracting an optimal basis to the problem of finding efficient features for representing speech signals of a given speaker The speech segments are assumed to be generated by a linear combination of the basis functions, thus the distribution of speech segments of a speaker is modeled by adapting the basis functions so that each source component is statistically independent. The learned basis functions are oriented and localized in both space and frequency, bearing a resemblance to Gabor wavelets. These features are speaker dependent characteristics and to assess their efficiency we performed speaker identification experiments and compared our results with the conventional Fourier-basis. Our results show that the proposed method is more efficient than the conventional Fourier-based features in that they can obtain a higher speaker identification rate.

• 음성과학
• /
• 제10권4호
• /
• pp.117-124
• /
• 2003

Speaker recognition is the technology that confirms the identification of speaker by using the characteristic of speech. Such technique is classified into speaker identification and speaker verification: The first method discriminates the speaker from the preregistered group and recognize the word, the second verifies the speaker who claims the identification. This method that extracts the information of speaker from the speech and confirms the individual identification becomes one of the most efficient technology as the service via telephone network is popularized. Some problems, however, must be solved for the real application as follows; The first thing is concerning that the safe method is necessary to reject the imposter because the recognition is not performed for the only preregistered customer. The second thing is about the fact that the characteristic of speech is changed as time goes by, So this fact causes the severe degradation of recognition rate and the inconvenience of users as the number of times to utter the text increases. The last thing is relating to the fact that the common characteristic among speakers causes the wrong recognition result. The silence parts being included the center of speech cause that identification rate is decreased. In this paper, to make improvement, We proposed identification rate can be improved by removing silence part before processing identification algorithm. The methods detecting speech area are zero crossing rate, energy of signal detect end point and starting point of the speech and process DTW algorithm by using two methods in this paper. As a result, the proposed method is obtained about 3% of improved recognition rate compare with the conventional methods.

• 대한음성학회지:말소리
• /
• 제47호
• /
• pp.121-129
• /
• 2003

This paper describes the use of pitch information for speaker identification. The recognition system is a GMM based one with 4 connected Korean digits speech database. The mean of the pitch period in voiced sections of speech are shown to be ,useful at discriminating between speakers. Utilizing this feature with Gaussian mixture model in the speaker identification system gave a marked improvement, maximum 6% improvement comparing to the baseline Gaussian mixture model.

• 음성과학
• /
• 제11권3호
• /
• pp.131-142
• /
• 2004

Speaker identification system basically functions by comparing spectral energy of an individual production model with that of an input signal. This study aimed to develop a new speaker identification system from two parameters from the spectral energy of numeric sounds: difference sum and correlation coefficient. A narrow-band spectrogram yielded more stable spectral energy across time than a wide-band one. In this paper, we collected empirical data from four male speakers and tested the speaker identification system. The subjects produced 18 combinations of three-digit numeric. sounds !en times each. Five productions of each three-digit number were statistically averaged to make a model for each speaker. Then, the remaining five productions were tested on the system. Results showed that when the threshold for the absolute difference sum was set to 1200, all the speakers could not pass the system while everybody could pass if set to 2800. The minimum correlation coefficient to allow all to pass was 0.82 while the coefficient of 0.95 rejected all. Thus, both threshold levels can be adjusted to the need of speaker identification system, which is desirable for further study.

• 한국음향학회지
• /
• 제22권6호
• /
• pp.457-462
• /
• 2003

본 논문에서는 GMM (Gaussian Mixture Model)에 기반한 문맥독립 화자식별 시스템의 식별성능 향상과 실시간 처리를 위한 계산량 감소를 위하여 화자 프루닝 (Speaker Pruning) 방법을 제안한다. 기존의 화자식별 방법인 최대유사도(Maximum Likelihood) 방법과 가중모델순위 (Weighting Model Rank) 방법, 수정된 가중모델순위 (Modified WMR) 방법 등은 입력 음성 전체와 모든 화자모델들과의 유사도를 프레임 단위로 계산하여 가장 큰 누적 유사도를 가지는 화자를 식별화자로 결정하는 방법으로써, 입력 프레임 및 등록 화자수가 늘어남에 따라 계산량 및 식별시간이 늘어나는 단점이 있었다. 이러한 단점을 해결하기 위하여, 제안방법은 입력음성 프레임의 일부분만을 이용하여 화자모델들과의 프레임 유사도를 계산한 후 계산된 유사도를 이용하여 등록화자의 상위 일부분의 화자만을 선택하고, 선택된 화자들에서만 유사도 계산을 수행함으로서 계산량 및 식별시간을 줄이는 방법이다. 또한, 화자 프루닝을 적용할 경우 화자수가 가변 되더라도 수정된 가중모델 순위방법을 적용할 수 있어 식별성능을 높일 수 있다. 식별실험결과, 제안방법을 적용한 경우 기존의 최대 유사도 방법이나 가중모델순위 방법보다 최대 65%의 계산량 및 식별시간을 감소시킬 수 있었으며, 약 2%의 향상된 식별결과를 나타내어, 본 논문에서 제안한 방법의 유효성을 확인할 수 있었다.

• 대한음성학회지:말소리
• /
• 제61호
• /
• pp.113-123
• /
• 2007

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.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제5권1호
• /
• pp.76-82
• /
• 2005

Speech signal has various features of speakers. This feature is extracted from speech signal processing. The speaker is identified by the speaker identification system. In this paper, we propose the speaker identification system that uses the incremental learning based on neural network. Recorded speech signal through the microphone is blocked to the frame of 1024 speech samples. Energy is divided speech signal to voiced signal and unvoiced signal. The extracted 12 orders LPC cpestrum coefficients are used with input data for neural network. The speakers are identified with the speaker identification system using the neural network. The neural network has the structure of MLP which consists of 12 input nodes, 8 hidden nodes, and 4 output nodes. The number of output node means the identified speakers. The first output node is excited to the first speaker. Incremental learning begins when the new speaker is identified. Incremental learning is the learning algorithm that already learned weights are remembered and only the new weights that are created as adding new speaker are trained. It is learning algorithm that overcomes the fault of neural network. The neural network repeats the learning when the new speaker is entered to it. The architecture of neural network is extended with the number of speakers. Therefore, this system can learn without the restricted number of speakers.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2005년도 추계 학술대회 발표논문집
• /
• pp.159-162
• /
• 2005

Identification is the process automatically identify who is speaking on the basis of information obtained from speech waves. In training phase, each speaker models are trained using each speaker's speech data. GMMs (Gaussian Mixture Models), which have been successfully applied to speaker modeling in text-independent speaker identification, are not efficient in insufficient training data environment. This paper proposes speaker modeling method using MLLR (Maximum Likelihood Linear Regression) method which is used for speaker adaptation in speech recognition. We make SD-like model using MLLR adaptation method instead of speaker dependent model (SD). Proposed system outperforms the GMMs in small training data environment.