• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.7
• /
• pp.1539-1544
• /
• 2003

Recently, many researchers have studied the speaker recognition through the statistical processing method using Karhonen-Loeve Transform. However, the content of speaker's identity and the vocalization speed cause speaker recognition rate to be lowered. This parer studies the speaker recognition method using weighted parameters which are weighted with eigen-values of speech so as to emphasize the speaker's identity and optimal path which is made by DWP so as to normalize dynamic time feature of speech. To confirm this method, we compare the speaker recognition rate from this proposed method with that from the conventional statistical processing method. As a result, it is shown that this method is more excellent in speaker recognition rate than conventional method.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.1
• /
• pp.47-50
• /
• 2019

This paper proposes an acoustic model adaptation method for effective speech recognition in noisy environments. In the proposed algorithm, the noise corruption function is estimated employing DNN (Deep Neural Network), and the function is applied to the model parameter estimation. The experimental results using the Aurora 2.0 framework and database demonstrate that the proposed model adaptation method shows more effective in known and unknown noisy environments compared to the conventional methods. In particular, the experiments of the unknown environments show 15.87 % of relative improvement in the average of WER (Word Error Rate).

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.7C
• /
• pp.598-603
• /
• 2010

In this paper, we work for Lipreading using visual information for ship environment. Lipreading is studied for using image information including lips of a speaker at the existing speech recognition system. This technique is a compensation method to increase recognition rate decreasing remarkably in noisy circumstances. Proposed way improved the rate of recognition improving methode of preprocessing using the Gabor Filter for Ship Environment. The experiment were carried out under changing of light with time in the ship environment with lip image. For Comparing with recognition, make a compare with between method of lip region of interest (ROI) before Gabor filtering and after Gabor filtering. In the case of using method of lip ROI before Gabor filtering, the result of the experiments applying to the proposed ways recognition resulting in 44% of recognition.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.5
• /
• pp.949-958
• /
• 2018

A voice recognition control system is a system for controlling a peripheral device by recognizing a voice. Recently, a voice recognition control system have been applied not only to smart devices but also to various environments ranging from IoT(: Internet of Things), robots, and vehicles. In such a voice recognition control system, the recognition rate is lowered due to the ambient noise in addition to the voice of the user. In this paper, we propose a dual channel acoustic beamforming hardware architecture for MEMS(: Microelectromechanical Systems) microphones to eliminate ambient noise in addition to user's voice. And the proposed hardware architecture is designed as ASIC(: Application-Specific Integrated Circuit) using TowerJazz $0.18{\mu}m$ CMOS(: Complementary Metal-Oxide Semiconductor) technology. The designed dual channel acoustic beamforming ASIC has a die size of $48mm^2$, and the directivity index of the user's voice were measured to be 4.233㏈.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.08a
• /
• pp.298-301
• /
• 1998

HMM 훈련 방법에 따른 음소인식률의 변화에 대하여 기술한다. 음성모델은 이산 확률 밀도 혹은 연속 확률 밀도를 갖는 HMM을 사용하였으며, 훈련 알고리즘으로서는 forward-backward 와 segmental K-means 알고리즘을 사용하였다. 연속 확률 밀도는 N개의 mixture로 구성되어 있는데 1개의 mixture로 확장할 경우에서는 이진 트리 방식과 one-by-one 방식을 사용하였다. 여러 가지의 조합을 이용하여 음소인식 실험을 수행한 결과 연속 확률 분포를 사용하고 one-by-one 방식을 사용한 forward-backward 알고리즘이 가장 우수한 결과를 나타내었다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 1998.10a
• /
• pp.226-231
• /
• 1998

본 연구에서는 짧은 시간에 계산이 가능하며, 음성을 특징 지울 수 있는 파라미터로서 영 교차율(zero crossing rate), 단 구간 에너지(short-term, energy) 그리고 포만트(formant)를 사용하였다. 특정 화자의 음성을 입력 받아서 단모음인 'ㅏ, ㅐ, ㅓ, ㅔ, ㅗ, ㅜ, ㅡ. ㅣ'에 대한 인식을 위해 위의 세가지 파라미터를 측정하였다. 영 교차율과 단 구간 에너지 파라미터는 유성음과 무성음의 구별과 음성인지 아닌지를 판별하는데 사용하였다. 포만트 파라미터는 10차 켑스트럼(cepstrum)을 이용하여 구하였으며, 각 단모음을 판별하기 위해서 사용하였다. 하나의 단모음을 입력받아 처리하여 텍스트로 출력하는데 평균 0.065sec에 처리하며, 각각의 단모음에 대해 93%, 10개의 테스트 문장에 대해 72%의 인식률을 보이고 있다.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.3
• /
• pp.61-70
• /
• 1995

In this paper, the PLP cepstrum which is close to human perceptual characteristics was extracted through the spread time area to get the temperal feature. Phonemes were recognized by artificial neural network similar to the learning method of human. The phoneme strings were matched by Markov models which well suited for sequence. Phoneme recognition for the continuous Korean speech had been done using speech blocks in which speech frames were gathered with unequal numbers. We parameterized the blocks using 7th order PLPs, PTP, zero crossing rate and energy, which neural network used as inputs. The 100 data composed of 10 Korean sentences which were taken from the speech two men pronounced five times for each sentence were used for the the recognition. As a result, maximum recognition rate of 94.4% was obtained. The sentence was recognized using Markov models generated by the phoneme strings recognized from earlier results the recognition for the 200 data which two men sounded 10 times for each sentence had been carried out. The sentence recognition rate of 92.5% was obtained.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.3
• /
• pp.273-278
• /
• 2004

In this paper, the recurrent neural networks (RNN) is applied to compensate for HMM recognition algorithm, which is commonly used as main recognizer. Among these recurrent neural networks, the multi-layer recurrent neural prediction model (MRNPM), which allows operating in real-time, is used to implement learning and recognition, and HMM and MRNPM are used to design a hybrid-type main recognizer. After testing the designed speech recognition algorithm with Korean number pronunciations (13 words), which are hardly distinct, for its speech-independent recognition ratio, about 5% improvement was obtained comparing with existing HMM recognizers. Based on this result, only optimal (recognition) codes were extracted in the actual DSP (TMS320C6711) environment, and the embedded speech recognition system was implemented. Similarly, the implementation result of the embedded system showed more improved recognition system implementation than existing solid HMM recognition systems.

• KIISE Transactions on Computing Practices
• /
• v.22 no.2
• /
• pp.69-76
• /
• 2016

With the recent increase of the interest in IoT in almost all areas of industry, computing technologies have been increasingly applied in human environments such as houses, buildings, cars, and streets; in these IoT environments, speech recognition is being widely accepted as a means of HCI. The existing server-based speech recognition techniques are typically fast and show quite high recognition rates; however, an internet connection is necessary, and complicated server computing is required because a voice is recognized by units of words that are stored in server databases. This paper, as a successive research results of speech recognition algorithms for the Korean phonemic vowel 'ㅏ', 'ㅓ', suggests an implementation of speech recognition algorithms for the Korean phonemic vowel 'ㅣ'. We observed that almost all of the vocal waveform patterns for 'ㅣ' are unique and different when compared with the patterns of the 'ㅏ' and 'ㅓ' waveforms. In this paper we propose specific waveform patterns for the Korean vowel 'ㅣ' and the corresponding recognition algorithms. We also presents experiment results showing that, by adding neural-network learning to our algorithm, the voice recognition success rate for the vowel 'ㅣ' can be increased. As a result we observed that 90% or more of the vocal expressions of the vowel 'ㅣ' can be successfully recognized when our algorithms are used.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06c
• /
• pp.378-382
• /
• 1994

HMM 세그멘테이션을 이용하여 LVQ 알고리즘에 적용시킨 하이브리드 음성인식에 관한 연구이다. LVQ 학습알고리즘은 정적 패턴 분리를 위한 참조벡터 즉, 고정차원인 벡터들을 생성하는데 유리하다. 하이브리드 알고리즘은 정적패턴 인식에 사용 되어지는 LVQ 알고리즘에 HMM 세그멘테이션을 접목시켜 입력패턴을 정규화된 의미있는 값으로서 바꾸어 사용하는데 있다. 한국어 음절중 8개 모음 아, 이, 우, 에, 오, 애, 어, 으를 추출하여 인식실험을 하였다. 인식률은 화자종속일 경우 코드북수 256개를 기준으로 LVQ1, LVQ2, LVQ3, OLVQ1 알고리즘순으로 91.7%, 91.8%, 91.1%의 인식률을 구했고 화자 독립의 경우는 83.4%, 83.9%, 86.8%, 85.3%의 인식률을 구했다.