• Journal of Korean language education
• /
• v.29 no.3
• /
• pp.25-53
• /
• 2018

This study examined the influence of lexical factors on verbal Eojeol recognition. To meet the goal, forty-five L2 Korean learners and twenty-two Korean native speakers took Eojeol decision tasks measured with the lexical factors such as 'number of strokes', 'number of consonants and vowels', 'number of syllables', 'number of morphemes', 'whole Eojeol frequency', 'root frequency', 'first-syllable-sharing frequency', and 'number of dictionary meanings.' As a result, 'whole Eojeol frequency' was the most effective factor to predict Eojeol recognition reaction time for native speakers and L2 learners, which supports the full-list model. Other lexical factors influencing Eojeol recognition reaction time in L2 learners were different following their proficiency level.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.374-378
• /
• 2022

대화형 관계 추출의 목표는 주어진 대화에서 두 개체 간의 관계를 식별하는 것이다. 대화 중에 화자는 개체 및 관계와 관련이 있는 단서인 트리거를 통해 특정 개체 간 관계를 식별하는 것에 힌트를 얻을 수 있다. 그러나 데이터에 대해 항상 트리거 정보가 존재하는 것이 아니므로 트리거를 활용해 성능을 향상시키는 것은 어렵다. 본 논문은 이 문제점을 해소하기 위해 대화, 개체, 관계 중심으로 트리거 생성 모델을 학습하고, 이를 통해 생성된 트리거를 대화형 관계 추출에 학습하여 관계 식별에 효과적인 성능 향상을 보이는 접근법을 제안한다. 제안하는 접근법은 대화형 관계 추출 태스크에서 기존 성능과 비교한 결과 Dev, Test에서 각각 F1 19.74%p, F1 15.53%p 의 성능 향상을 보였다.

• Journal of Broadcast Engineering
• /
• v.20 no.1
• /
• pp.164-170
• /
• 2015

This paper deals with the adaptation of classification model in the binary mask approach to suppress noise in the noisy environment. The binary mask estimation approach is known to improve speech intelligibility of noisy speech. However, the same type of noisy data for the test data should be included in the training data for building the classification model of binary mask estimation. The eigenvoice adaptation is applied to the noise-independent classification model and the adapted model is used as noise-dependent model. The results are reported in Hit rates and False alarm rates. The experimental results confirmed that the accuracy of classification is improved as the number of adaptation sentences increases.

• Annual Conference on Human and Language Technology
• /
• 1989.10a
• /
• pp.101-106
• /
• 1989

음성 및 문자를 통한 컴퓨터와의 정보 교환을 위한 통합 사용자 인터페이스 (Intelligent Man- Machine interface) 시스템의 일환으로 한국어 단모음의 인식을 위한 시스템을 인공 신경망 모델을 사용하여 구현하였으며 인식시스템의 상위 접속부에 필요한 단어 인식 모듈에 있어서의 인지 실험도 행하였다. 모음인식의 입력으로는 제1, 제2, 제3 포르만트가 사용되었으며 실험대상은 한국어의 [아, 어, 오, 우, 으, 이, 애, 에]의 8 개의 단모음으로 하였다. 사용한 인공 신경망 모델은 Multilayer Perceptron 이며, 학습 규칙은 Generalized Delta Rule 이다. 1 인의 남성 화자에 대하여 약 94%의 인식율을 나타내었다. 그리고 음성 인식시의 인지 현상 실험을 위하여 약 20개의 단어를 인공신경망의 어휘레벨에 저장하여 음성의 왜곡, 인지시의 lexical 영향, categorical percetion등을 실험하였다. 이때의 인공 신경망 모델은 Interactive Activation and Competition Model을 사용하였으며, 음성 입력으로는 가상의 음성 피쳐 데이타를 사용하였다.

• MALSORI
• /
• no.65
• /
• pp.93-103
• /
• 2008

In this paper, we propose a method for optimizing a multiple pronunciation dictionary used for modeling pronunciation variations of non-native speech. The proposed method removes some confusable pronunciation variants in the dictionary, resulting in a reduced dictionary size and less decoding time for automatic speech recognition (ASR). To this end, a confusability measure is first defined based on the Levenshtein distance between two different pronunciation variants. Then, the number of phonemes for each pronunciation variant is incorporated into the confusability measure to compensate for ASR errors due to words of a shorter length. We investigate the effect of the proposed method on ASR performance, where Korean is selected as the target language and Korean utterances spoken by Chinese native speakers are considered as non-native speech. It is shown from the experiments that an ASR system using the multiple pronunciation dictionary optimized by the proposed method can provide a relative average word error rate reduction of 6.25%, with 11.67% less ASR decoding time, as compared with that using a multiple pronunciation dictionary without the optimization.

• Journal of KIISE:Computing Practices and Letters
• /
• v.13 no.5
• /
• pp.338-342
• /
• 2007

In this paper we propose a user adaptive post-processing method to improve the accuracy of speaker dependent, isolated word speech recognition, particularly for mobile devices. Our method considers the recognition result of the basic recognizer simply as a high-level speech feature and processes it further for correct recognition result. Our method learns correlation between the output of the basic recognizer and the correct final results and uses it to correct the erroneous output of the basic recognizer. A multi-layer perceptron model is built for each incorrectly recognized word with high frequency. As the result of experiments, we achieved a significant improvement of 41% in recognition accuracy (41% error correction rate).

• Phonetics and Speech Sciences
• /
• v.7 no.1
• /
• pp.3-10
• /
• 2015

In this paper, different frequency scales in cepstral feature extraction are evaluated for the text-independent speaker recognition. To this end, mel-frequency cepstral coefficients (MFCCs), linear frequency cepstral coefficients (LFCCs), and bilinear warped frequency cepstral coefficients (BWFCCs) are applied to the speaker recognition experiment. In addition, the spectro-temporal features extracted by the cepstral-time matrix (CTM) are examined as an alternative to the delta and delta-delta features. Experiments on the NIST speaker recognition evaluation (SRE) 2004 task are carried out using the Gaussian mixture model-universal background model (GMM-UBM) method and the joint factor analysis (JFA) method, both based on the ALIZE 3.0 toolkit. Experimental results using both the methods show that BWFCC with appropriate warping factor yields better performance than MFCC and LFCC. It is also shown that the feature set including the spectro-temporal information based on the CTM outperforms the conventional feature set including the delta and delta-delta features.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.8
• /
• pp.590-596
• /
• 2004

This paper Presents a new self-adaptation algorithm based on maximum a posteriori (MAP) eigenvoice for Korean connected digit recognition. The proposed MAP eigenvoice is developed by introducing a probability density model for the eigenvoice coefficients. The Proposed approach provides a unified framework that incorporates the Prior model into the conventional eigenvoice estimation. In self-adaptation system we use only one adaptation utterance that will be recognized, we use MAP eigenvoice that is most robust adaptation. In series of self-adaptation experiments on the Korean connected digit recognition task. we demonstrate that the performance of the proposed approach is better than that of the conventional eigenvoice algorithm for a small amount of adaptation data.

• 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.

• Phonetics and Speech Sciences
• /
• v.4 no.1
• /
• pp.91-98
• /
• 2012

The purposes of this paper is to study the characteristics of compared to the speakers voice without depression and speakers with depression, and to propose a objective method for the measurement of the therapeutic effects as well as for diagnostics of depression based on the characteristics. The voice samples obtained from 11 female speakers with depression, aged from 20 to 40, diagnosed as having major depressive disorder by an psychiatrist were compared with those from 12 normal controls with matched sex, age, height, weight, education, smoking, and drinking. The voice samples are taken by a portable digital recorder(TASCAM DR-07, Japan) and analysed using the MDVP(Multi-Dimentional Voice Program) software module from CSL(Computerized Speech Lab, kay elemetrics, co, model 4100). The result of the investigation are as following. First, the average speaking fundamental frequency and loudness range of the speakers with depression group was statistically significantly lower than that of the control group. The pitch range of the control group was rather higher than that of the speakers with depression group, but without statistical significance. Overall speech rates have no statistical difference between two groups. Second, the average speaking fundamental frequency and loudness range have statistically significant negative correlation with Beck Depression Inventory, i. e. more severe depression exhibits lower average speaking fundamental frequency and loudness range. Other vocal parameters such as pitch range and overall speech rate have no statistically meaningful correlations with Beck Depression Inventory.