• The Journal of the Acoustical Society of Korea
• /
• v.13 no.3
• /
• pp.63-70
• /
• 1994

In this Paper, we propose a new codeword tying algorithm based on a tree structured classfier. The proposed algorithm which can be viewed as a kind of soft decision using statistical properties between codewords and states has an advantage of fast construction, and guarantees a unique optimal solution. Also, it can easily be applied to any speech recognition system based on discrete hidden Markov model (HMM). Experimental results on speaker-independent isolated word recognition show error reduction of $6\%$ for the codebook of size 256 and $9\%$ for 512 size and also HMM parameter reduction of about $20\%$.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.5 no.4
• /
• pp.363-371
• /
• 2010

In an effort to enhance the quality of feature vector classification and thereby reduce the recognition error rate for the speaker-independent speech recognition, we study the effect of spectral tilt on the Fourier magnitude spectrum en route to the extraction of MFCC. The effect of FIR filtering on the speech signal on the speech recognition is also investigated in parallel. Evaluation of the proposed methods are performed by two independent ways of the Fisher discriminant objective function and speech recognition test by hidden Markov model with fuzzy vector quantization. From the experiments, the recognition error rate is found to show about 10% relative improvements over the conventional method by an appropriate choice of the tilt factor.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.9C
• /
• pp.907-912
• /
• 2005

An efficient method for calculation of observation probability in CDHMM(Continous Density Hidden Markov Model) is proposed in this paper. the proposed algorithm, called FCOP(Fast Computation of Observation Probability), approximate obsewation probabilities in CDHMM by eliminating insignificant PDFs(Probability Density Functions) and reduces the computational load. When applied to a speech recognition system, the proposed FCOP algorithm can reduce the instruction cycles by $20\%-30\%$ and can also increase the recognition speed about $30\%$ while minimizing the loss in its recognition rate. When implemented on a practical cellular phone, the FCOP algorithm can increase its recognition speed about $30\%$ while suffering $0.2\%$ loss in recognition rate.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.3
• /
• pp.475-483
• /
• 2012

The characteristic features of speech signal do not vary significantly from frame to frame. Therefore, it is advisable to reduce the redundancy involved in the similar feature vectors. The objective of this paper is to search for the optimal condition of minimum redundancy and maximum relevancy of the speech feature vectors in speech recognition. For this purpose, we realize redundancy reduction by way of a vigilance parameter and investigate the resultant effect on the speaker-independent speech recognition of isolated words by using FVQ/HMM. Experimental results showed that the number of feature vectors might be reduced by 30% without deteriorating the speech recognition accuracy.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.5
• /
• pp.68-75
• /
• 1997

Spectral subtraction is widely-used preprocessing technique for speech recognition in additive noise environments, but it requires a good estimate of the noise power spectrum. In this paper, we employ the histogram technique for the estimation of noise spectrum. This technique has advantages over other noise estimation methods in that it does not requires speech/non-speech detection and can estimate slowly-varying noise spectra. According to the speaker-independent isolated word recognition in both colored Gaussian and car noise environments under various SNR conditions. Histogram-technique-based spectral subtraction method yields superier performance to the one with conventional noise estimation method using the spectral average of initial frames during non-speech period.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.1E
• /
• pp.89-95
• /
• 1994

This paper describes development of a real-time voice dialing system which can recognize around one hundred word vocabularies in speaker independent mode. The voice recognition algorithm in this system is implemented on a DSP board with a telephone interface plugged in an IBM PC AT/486. In the DSP board, procedures for feature extraction, vector quantization(VQ), and end-point detection are performed simultaneously in every 10 msec frame interval to satisfy real-time constraints after detecting the word starting point. In addition, we optimize the VQ codebook size and the end-point detection procedure to reduce recognition time and memory requirement. The demonstration system has been displayed in MOBILAB of the Korean Mobile Telecom at the Taejon EXPO'93.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.4
• /
• pp.224-229
• /
• 2005

In this paper, we Propose a novel cepstrum normalization method which normalizes the probability density function (pdf) of cepstrum for robust speech recognition in additive noise environments. While the conventional methods normalize the first- and/or second-order statistics such as the mean and/or variance of the cepstrum. the proposed method fully normalizes the statistics of cepstrum by making the pdfs of clean and noisy cepstrum identical to each other For the target Pdf, the generalized Gaussian distribution is selected to consider various densities. In recognition phase, we devise a table lookup method to save computational costs. From the speaker-independent isolated-word recognition experiments, we show that the Proposed method gives improved Performance compared with that of the conventional methods, especially in heavy noise environments.

• Annual Conference on Human and Language Technology
• /
• 1992.10a
• /
• pp.185-192
• /
• 1992

이 논문에서는 음성인식기술을 이용한 연구센타 부서안내 시스팀(KARS:Korea Telecom Automatic voice Recognition system)에 대하여 기술하였다. 이 시스팀은 기본적으로 음성응답 시스팀과 유사하지만 명령입력을 위해 푸시버튼 대신 음성을 이용한다는 점이 다르다. 사용자가 마이크로폰을 통해 음성명령을 입력하면, 이 시스팀은 사용자의 음성명령을 인식하여 연구센타내 각 부서의 간략한 소개, 전화번호 및 위치를 안내해 준다. 이 시스팀은 HMM(Hidden Markov Model)을 이용하는 화자독립 격리단어 인식시스팀으로서 116개의 부서이름과 7개의 제어용 단어로 구성되어 있는 123개 단어를 인식할 수 있다. 이 시스팀은 음소와 유사한 한국어 서브워드(subword)를 HMM의 기본단위로 사용하며 인식 실험결과 98.6%의 인식율을 얻을 수 있었다.

• The Journal of the Acoustical Society of Korea
• /
• v.6 no.3
• /
• pp.60-67
• /
• 1987

In this paper, a phoneme classification method for Korean speech recognition has been proposed and its performance has been studied. The phoneme classification has been done based on the phonemic features extracted by the prewindowed recursive least-square (PRLS) algorithm that is a kind of adaptive filter algorithms. Applying the PRLS algorithm to input speech signal, precise detection of phoneme boundaries has been made, Reference patterns of Korean phonemes have been generated by the ordinery vector quantization (VQ) of feature vectors obtained manualy from prototype regions of each phoneme. In order to obtain the performance of the proposed phoneme classification method, the method has been tested using spoken names of seven Korean cities which have eleven different consonants and eight different vowels. In the speaker-dependent phoneme classification, the accuracy is about $85\%$ considering simple phonemic rules of Korean language, while the accuracy of the speaker-independent case is far less than that of the speaker-dependent case.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.6
• /
• pp.5-14
• /
• 1994

This paper is about an experiment of speaker-independent automatic Korean spoken words recognition using Multi-Layered Perceptron and Error Back-propagation algorithm. The object words are 50 citynames of D.D.D local numbers. 43 of those are 2 syllables and the rest 7 are 3 syllables. The words were not segmented into syllables or phonemes, and some feature components extracted from the words in equal gap were applied to the neural network. That led independent result on the speech duration, and the PARCOR coefficients calculated from the frames using linear predictive analysis were employed as feature components. This paper tried to find out the optimum conditions through 4 differerent experiments which are comparison between total and pre-classified training, dependency of recognition rate on the number of frames and PAROCR order, recognition change due to the number of neurons in the hidden layer, and the comparison of the output pattern composition method of output neurons. As a result, the recognition rate of $89.6\%$ is obtaimed through the research.