• 한국음향학회지
• /
• 제18권1호
• /
• pp.86-91
• /
• 1999

본 연구에서는 문맥 종속 또는 문맥 독립형 화자 인식에서의 단점을 개선하는 방법으로 문맥 제시형 화자 인식 실험을 수행하였다. 화자 인식 알고리즘으로는 개선된 Dynamic Time Warping(DTW)을 사용하였고 실시간 처리를 위하여 전체 계산량을 증가시키지 않는 아주 간단한 끝점검출알고리즘을 사용하였으며, 여러 가지 다양한 특징 파라미터를 이용하여 인식실험을 행한 결과 weighted cepstrum을 이용했을 때 가장 좋은 인식성능을 얻을 수 있었다. 실험결과 세 개의 단어를 제시하였을 경우 화자식별오류는 0.02%를 보였고, 화자확인은 문턱값을 적절히 정했을 때 사용자 거부율 1.89%, 사칭자 허용률 0.77%, 총 확인 오류0.97%를 보였다.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1994년도 제11회 음성통신 및 신호처리 워크샵 논문집 (SCAS 11권 1호)
• /
• pp.358-363
• /
• 1994

This paper presents the improvement of the SCHMM using discrete VQ and One-Pass algorithm for keywords recognition in Korean continuous speech. The SCHMM using discrete VQ is a simple model that is composed of a variable mixture gaussian probability density function with dynamic mixture number. One-Pass algorithm is improved such that recognition rates are enhanced by fathoming any undesirable semisyllable with the low likelihood and the high duration penalty, and computation time is reduced by testing only the frame which is dissimilar to the previously testd frame. In recognition experiments for speaker-dependent case, the improved One-Pass algorithm has shown recognition rates as high as 99.7% and has reduced compution time by about 30% compared with the currently abailable one-pass algorithm.

• 한국지능시스템학회논문지
• /
• 제14권2호
• /
• pp.201-206
• /
• 2004

This study attempted to develop a speech recognition module applied to a wheelchair for the physically handicapped. In the proposed speech recognition module, TMS320C32 was used as a main processor and Mel-Cepstrum 12 Order was applied to the pro-processor step to increase the recognition rate in a noisy environment. DTW (Dynamic Time Warping) was used and proven to be excellent output for the speaker-dependent recognition part. In order to utilize this algorithm more effectively, the reference data was compressed to 1/12 using vector quantization so as to decrease memory. In this paper, the necessary diverse technology (End-point detection, DMA processing, etc.) was managed so as to utilize the speech recognition system in real time

• 음성과학
• /
• 제15권2호
• /
• pp.67-77
• /
• 2008

In this paper, we implemented a robust speech recognizer using the TMS320VC33 DSP. For this implementation, we had built speech and noise database suitable for the recognizer using spectral subtraction method for noise removal. The recognizer has an explicit structure in aspect that a speech signal is enhanced through spectral subtraction before endpoints detection and feature extraction. This helps make the operation of the recognizer clear and build HMM models which give minimum model-mismatch. Since the recognizer was developed for the purpose of controlling car facilities and voice dialing, it has two recognition engines, speaker independent one for controlling car facilities and speaker dependent one for voice dialing. We adopted a conventional DTW algorithm for the latter and a continuous HMM for the former. Though various off-line recognition test, we made a selection of optimal conditions of several recognition parameters for a resource-limited embedded recognizer, which led to HMM models of the three mixtures per state. The car noise added speech database is enhanced using spectral subtraction before HMM parameter estimation for reducing model-mismatch caused by nonlinear distortion from spectral subtraction. The hardware module developed includes a microcontroller for host interface which processes the protocol between the DSP and a host.

• 전자공학회논문지B
• /
• 제32B권11호
• /
• pp.1489-1495
• /
• 1995

In this paper, we propose recurrent neural prediction models (RNPM), recurrent neural networks trained as a nonlinear predictor of speech, as a new connectionist model for speech recognition. RNPM modulates its mapping effectively by internal representation, and it requires no time alignment algorithm. Therefore, computational load at the recognition stage is reduced substantially compared with the well known predictive neural networks (PNN), and the size of the required memory is much smaller. And, RNPM does not suffer from the problem of deciding the time varying target function. In the speaker dependent and independent speech recognition experiments under the various conditions, the proposed model was comparable in recognition performance to the PNN, while retaining the above merits that PNN doesn't have.

• 대한전자공학회논문지
• /
• 제24권5호
• /
• pp.905-913
• /
• 1987

This paper presents a speaker dependent isolated digit recognition algorithm using the Mellin transform. Since the Mellin transform converts a scale information into a phase information, attempts have been made to utilize this scale invariance property of the Mellin transform in order to alleviate a time-normalization procedure required for a speech recognition. It has been found that good results can be obtained by taking the Mellin transform to the features such as a ZCR, log energy, normalized autocorrelation coefficients, first predictor coefficient and normalized prediction error. We employed a difference function for evaluating a similarity between two patterns. When the proposed algorithm was tested on Korean digit words, a recognition rate of 83.3% was obtained. The recognition accuracy is not compatible with the other technique such as LPC distance however, it is believed that the Mellin transform can effectively perform the time-normalization processing for the speech recognition.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 추계학술대회 논문집 학회본부
• /
• pp.594-596
• /
• 1997

This paper presents the speaker-dependent Korean isolated digit recognition system using advanced competitive learning. Since competitive learning algorithms are easy and simple to implement, they are used in various fields. The proposed recognition algorithm consists of three procedures: comparing winning number of codebook vectors, selecting the representative vector out of codebook vectors, and generating a new codebook with the representative vectors. In this paper, we use a sound blaster 16 for obtaining speech data. Speech data are sampled by 16 bits and 11 kHz sampling rate.

• 한국음향학회지
• /
• 제14권5호
• /
• pp.74-82
• /
• 1995

본 논문은 HMM과 연결 숫자음의 후처리를 이용한 음성 다이얼링에 관한 연구이다. HMM(Hidden Markov Model)은 좋은 결과를 보이면서 현재 음성 인식 분야에서 널리 사용되는 알고리즘이다. 그러나, HMM의 학습 방법인 maximum like-lihood estimation은 인식률을 극대화하는 모델의 파라메터 값을 생성하지 못하는 단점이 었다. 이러한 문제점을 보완하기 위하여 Segmental K-means 학습 과정에 후저리를 이용하여 인식 실험을 하였다. 한국어 연속 숫자음은 영어 연속 숫자음과 달리 연음 현상의 영향을 많이 받는다. Level Building 과정에서 연음에 의한 오류를 감소시키기 위해 연음에 의해 발생할 수 있는 단어를 별도의 모델로 추가하였다. 이렇게 추가된 단어 모델들에 대한 몇 가지 규칙을 인식 결과에 적용하여 출력을 다시 조정한다. 본 시 스템은 TMS320C30 프로세서를 내장한 DSP 보드와 IBM PC 상에서 구현되었고, 표준 패턴은 실험실 잡음 환경에서 남성 화자3명을 대상으로 작성하였다. 인식 실험 결과 21종 전화 번호 252개 데이타에 대하여 화자 종속의 경우 $91.6\%$, 회자 독립의 경우 $80.5\%$의 인식률을 나타내었다.

• 한국음향학회지
• /
• 제14권3호
• /
• pp.97-104
• /
• 1995

본 논문은 음성 다이얼링 시스템 구현을 위한 연속 숫자음 인식에 관한 연구로써, 구문 분석을 이용한 One-Stage DP에 의한 음성 인식 방법을 제안하다. 인식 실험을 위해 우선 구간 구분화 알고리즘을 이용하여 DMS (Dynamic Multi-SEction) 모델을 만들며, 제안된 구문 분석을 이용한 One-Stage DP 방법으로 실험 대ㅛ상의 연속 숫자음 데이터를 인식하게 하였다. 본 연구에서는 8명의 ㅣ남성 화자에 의해 2-3번 발음도니 21종의 7자리의 연속 숫자음이 사용되었고, 기존의 One-Stage DP와 제안된 구문 분석을 이용한 One-Stage DP 음성 인식 알고리즘을 사용해서 화자 종속과 화자 독립 실험을 실험실 환경에서 수행하였다. 인식 실험 결과, 기존의 방법보다 제안된 방법이 인식률이 좋은 것으로 나타났으며, 제안된 방법에서는 화자 종속과 화자 독립 실험에서 각각 약 91.7%, 89.7%로 나타났다.

• 전자공학회논문지B
• /
• 제30B권2호
• /
• pp.51-58
• /
• 1993

This paper describes the development of a speaker-dependent isolated word recognizer as applied to voice dialing in a car noise environment. for this purpose, several methods to improve performance under such condition are evaluated using database collected in a small car moving at 100km/h The main features of the recognizer are as follow: The endpoint detection error can be reduced by using the magnitude of the signal which is inverse filtered by the AR model of the background noise, and it can be compensated by using variants of the DTW algorithm. To remove the noise, an autocorrelation subtraction method is used with the constraint that residual energy obtainable by linear predictive analysis should be positive. By using the noise rubust distance measure, distortion of the feature vector is minimized. The speech recognizer is implemented using the Motorola DSP56001(24-bit general purpose digital signal processor). The recognition database is composed of 50 Korean names spoken by 3 male speakers. The recognition error rate of the system is reduced to 4.3% using a single reference pattern for each word and 1.5% using 2 reference patterns for each word.