• 말소리와 음성과학
• /
• 제6권2호
• /
• pp.29-34
• /
• 2014

In the Vector Taylor Series (VTS)-based noisy speech recognition methods, Hidden Markov Models (HMM) are usually trained with clean speech. However, better performance is expected by training the HMM with noisy speech. In a previous study, we could find that Minimum Mean Square Error (MMSE) estimation of the training noisy speech in the log-spectrum domain produce improved recognition results, but since the proposed algorithm was done in the log-spectrum domain, it could not be used for the HMM adaptation. In this paper, we modify the previous algorithm to derive a novel mathematical relation between test and training noisy speech in the cepstrum domain and the mean and covariance of the Multi-condition TRaining (MTR) trained noisy speech HMM are adapted. In the noisy speech recognition experiments on the Aurora 2 database, the proposed method produced 10.6% of relative improvement in Word Error Rates (WERs) over the MTR method while the previous MMSE estimation of the training noisy speech produced 4.3% of relative improvement, which shows the superiority of the proposed method.

• 디지털콘텐츠학회 논문지
• /
• 제18권6호
• /
• pp.1049-1056
• /
• 2017

본 연구는 스피치교육의 효과를 측정하는데 있어 화자의 목소리의 변화에 주목하였다. 본 연구에서는 스피치교육을 통해 얻게 되는 실질적 효과 중 목소리의 변화를 보다 가시적이고, 과학적으로 평가하고자 하였다. 연구결과 모든 학습자의 목소리에서 스피치교육 전과는 다른 객관적인 변화를 찾을 수 있었다. 학습자 모두 공명, 음색, 발음의 정확성, 휴지 등 다양한 목소리 요소에서 점진적 기술향상이 이루어졌다. 즉, 스피치교육을 받기 전보다 목소리가 풍부해지고 발음이 정확하고, 휴지를 잘 활용하는 안정화된 결과를 볼 수 있었다. 이 연구결과를 통해 스피치훈련을 통해 목소리의 변화가 나타날 수 있는지 분석하고, 스피치 학습자들이 스피치교육에 적극 임해 스피치실력 향상의 결과를 얻을 수 있을 것으로 기대된다.

• 대한전기학회논문지
• /
• 제41권2호
• /
• pp.209-216
• /
• 1992

Deafs train articulation by observing mouth of a tutor, sensing tactually the motions of the vocal organs, or using speech training aids. Present speech training aids for deafs can measure only single speech parameter, or display only frequency spectra in histogram of pseudo-color. In this study, a speech training aids that can display subject's articulation in the form of a cross section of the vocal organs and other speech parameters together in a single system is to be developed and this system makes a subject know where to correct. For our objective, first, speech production mechanism is assumed to be AR model in order to estimate articulatory motions of the vocal organs from speech signal. Next, a vocal tract profile model using LP analysis is made up. And using this model, articulatory motions for Korean vowels are estimated and displayed in the vocal tract profile graphics.

• 인터넷정보학회논문지
• /
• 제2권1호
• /
• pp.97-106
• /
• 2001

본 논문에서는 발성장애아의 음성적 특징을 중심으로 컴퓨터 기반 발성훈련시스템을 설계 및 구현하였다. 본 발성훈련시스템은 선행훈련, 발성인지훈련, 발성강화훈련 단계로 구성되어 있으며, 발성장애 아동의 발성의 상황과 레벨을 분석하고 반복학습 및 개별학습이 가능하도록 하였다. 컴퓨터를 기반으로 발성장애아의 음성을 디지털 신호처리하기 위해 음성적 파라미터 즉, 음성의 강도, 음성의 고저, 유 무성음을 추출하였다. 추출된 음성적 파라미터는 이동체의 움직임 벡터 값으로 변환하여 이미지, 애니메이션, 게임적 요소와 같이 시각적으로 피드백 할 수 있도록 하였다.

• 말소리와 음성과학
• /
• 제2권3호
• /
• pp.29-36
• /
• 2010

The purposes of this study are (1) to see if language transfer effect is found in Korean speakers' pronunciation of English stops and to correct them and (2) to investigate the effectiveness of mimicry training and Speech Analyzer training on subjects' pronunciation of English stops. For these purposes, 20 Korean speakers' VOT values of English stops were measured using Speech Analyzer and their post-training production was compared with their pre-training production. The result shows that Korean speakers have no difficulty in correcting pronunciation errors of English voiceless stops and voiced stops and such a result indicates that language transfer effect is not noticed as expected. In addition, the result of pronunciation training shows that the training using Speech Analyzer is more effective than mimicry training.

• 음성과학
• /
• 제11권4호
• /
• pp.43-52
• /
• 2004

Model parameters in HMM based speech recognition systems are normally estimated using Maximum Likelihood Estimation(MLE). The MLE method is based mainly on the principle of statistical data fitting in terms of increasing the HMM likelihood. The optimality of this training criterion is conditioned on the availability of infinite amount of training data and the correct choice of model. However, in practice, neither of these conditions is satisfied. In this paper, we propose a training algorithm, MCE(Minimum Classification Error), to improve the performance of a speech recognizer detecting mispronunciation of a foreign language. During the conventional MLE(Maximum Likelihood Estimation) training, the model parameters are adjusted to increase the likelihood of the word strings corresponding to the training utterances without taking account of the probability of other possible word strings. In contrast to MLE, the MCE training scheme takes account of possible competing word hypotheses and tries to reduce the probability of incorrect hypotheses. The discriminant training method using MCE shows better recognition results than the MLE method does.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1991년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 22-24 Oct. 1991
• /
• pp.746-751
• /
• 1991

Deafs train articulation by observing mouth of a tutor. sensing tactually the notions of the vocal organs, or using speech training aids. Present speech training aids for deafs can measure only single speech ter, or display only frequency spectra in histogrm or pseudo-color. In this study, a speech training aids that can display subject's articulation in the form of a cross section of the vocal organs and other speech parameters together in a single system Is aimed to develop and this system makes a subject to know where to correct. For our objective, first, speech production mechanism is assumed to be AR model in order to estimate articulatory notions of the vocal tract from speech signal. Next, a vocal tract profile mode using LPC analysis is made up. And using this model, articulatory notions for Korean vowels are estimated and displayed in the vocal tract profile graphics.

• Journal of Audiology & Otology
• /
• 제25권1호
• /
• pp.22-26
• /
• 2021

Background and Objectives: Music listening has a concomitant effect on structural and functional organization of the brain. It helps in relaxation, mind training and neural strengthening. In relation to it, the present study was aimed to find the effect of Carnatic music listening training (MLT) on speech in noise performance in adults. Subjects and Methods: A total of 28 participants (40-70 years) were recruited in the study. Based on randomized control trial, they were divided into intervention and control group. Intervention group underwent a short-term MLT. Quick Speech-in-Noise in Kannada was used as an outcome measure. Results: Results were analysed using mixed method analysis of variance (ANOVA) and repeated measures ANOVA. There was a significant difference between intervention and control group post MLT. The results of the second continuum revealed no statistically significant difference between post training and follow-up scores in both the groups. Conclusions: In conclusion short-term MLT resulted in betterment of speech in noise performance. MLT can be hence used as a viable tool in formal auditory training for better prognosis.

• 대한청각학회지
• /
• 제25권1호
• /
• pp.22-26
• /
• 2021

Background and Objectives: Music listening has a concomitant effect on structural and functional organization of the brain. It helps in relaxation, mind training and neural strengthening. In relation to it, the present study was aimed to find the effect of Carnatic music listening training (MLT) on speech in noise performance in adults. Subjects and Methods: A total of 28 participants (40-70 years) were recruited in the study. Based on randomized control trial, they were divided into intervention and control group. Intervention group underwent a short-term MLT. Quick Speech-in-Noise in Kannada was used as an outcome measure. Results: Results were analysed using mixed method analysis of variance (ANOVA) and repeated measures ANOVA. There was a significant difference between intervention and control group post MLT. The results of the second continuum revealed no statistically significant difference between post training and follow-up scores in both the groups. Conclusions: In conclusion short-term MLT resulted in betterment of speech in noise performance. MLT can be hence used as a viable tool in formal auditory training for better prognosis.

• 음성과학
• /
• 제8권3호
• /
• pp.173-184
• /
• 2001

The purpose of this study was to examine the effects of manipulating loudness and pitch in terms of speech intelligibility and voice of a patient with Parkinson's Disease. The subject, who was diagnosed as a patient with Parkinson's disease 11 years ago, demonstrated a severely breath voice with low intensity. The accuracy of articulation in consonants was intelligible only at the single word level, and the overall intelligibility in continuous speech was low. The results showed that the subject's articulation accuracy and speech intelligibility was significantly improved after having loudness and pitch training. Habitual Fo, Jitter, Shimmer, Fo tremor, Amp tremor were decreased after training. In addition, the value of HNR also increased after training. It was shown that the changes of these acoustic parameters were closely related to the decrease of breathiness in Parkinson's voice, and this decrease of breathiness affected speech intelligibility considerably. Based on the experimental results, it was claimed that the vocal training by manipulating the loudness and pitch could be highly effective in improving the voice quality and speech intelligibility in Parkinson's Disease.