• Proceedings of the KOSOMBE Conference
• /
• v.1998 no.11
• /
• pp.161-162
• /
• 1998

In this paper, a speech processor which is the most important part of the cochlear implant is developed, to recover auditory ability for the sensorineural disorders who have damaged for their inner ear. This system consists of the analog and digital signal processing part, of which functions is the pre-processing and the main processing, respectively. The main processing is peformed in DSP processor (TMS320C31-40) by using S/W. Because the program is used in this system, it is possible to cope with the individual status of the patients, very easily.

• Speech Sciences
• /
• v.9 no.1
• /
• pp.89-110
• /
• 2002

Intonation is very difficult to describe and it is furthermore difficult to compare intonation between different languages because of their differences of intonation systems. This paper aims to compare some intonation phenomena between Korean, French and English. In this paper I will refer to ToBI (the Tone and Break Indices) which is a prosodic transcription model proposed originally by Pierrehumbert (1980) as a description tool. In the first part, I will summarize different ToBI systems, namely, K-ToBI (Korean ToBI), F-ToBI (French ToBI) and ToBI itself (English ToBI) in order to compare the differences of three languages within prosody. In the second part, I will analyze some tokens registered by Korean, French and American in different languages to show the difficulties of learning other languages and to find the prosodic cues to pronounce correctly other languages. The point of comparison in this study is the Accentual Phrase (AP) in Korean and in French and the intermediate phrase (ip) in English, which I will call ' subject phrase ' in this study for convenience.

• Phonetics and Speech Sciences
• /
• v.1 no.4
• /
• pp.97-103
• /
• 2009

This paper describes a new algorithm for encoding spectral envelope in the time domain alias cancellation (TDAC) part of G.729.1. The spectral envelope and modified discrete cosine transform (MDCT) coefficients of the weighted code-excited linear predictive (CELP) coding error in lower-band and the higher-band input signal are encoded in the TDAC part. In order to reduce allocation bits for spectral envelope coding, a new algorithm using sub-band correlation between adjacent frames is proposed. In addition, to improve the quality of decoded signals, two bit allocation strategies using reduced bits from the proposed algorithm are proposed. The performance of the proposed algorithm is evaluated in terms of objective quality and bit reduction rates. Experimental results show that the proposed algorithm increases the quality of sounds significantly.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.2E
• /
• pp.8-13
• /
• 2000

In this paper, we present an adaptive formant tracking algorithm for speech using closed phase WRLS-VFF-VT method. The pitch synchronous closed phase methods is known to give more accurate estimates of the vocal tract parameters than the pitch asynchronous method. However the use of a pitch-synchronous closed phase analysis method has been limited due to difficulties associated with the task of accurately isolating the closed phase region in successive periods of speech. Therefore we have implemented the pitch synchronous closed phase WRLS-VFF-VT algorithm for speech analysis, especially for formant tracking. The proposed algorithm with the variable threshold(VT) can provide a superior performance in the boundary of phone and voiced/unvoiced sound. The proposed method is experimentally compared with the other method such as two channel CPC method by using synthetic waveform and real speech data. From the experimental results, we found that the block data processing techniques, such as the two-channel CPC, gave reasonable estimates of the formant/antiformant. However, the data windows used by these methods included the effects of the periodic excitation pulses, which affected the accuracy of the estimated formants. On the other hand the proposed WRLS-VFF-VT method, which eliminated the influence of the pulse excitation by using an input estimation as part of the algorithm, gave very accurate formant/bandwidth estimates and good spectral matching.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06b
• /
• pp.16-18
• /
• 2006

한국어 형태소분석 및 품사부착에서 일부 품사는 높은 중의성으로 인하여 오류가 많으며, 일부 품사가 전체 오류의 대부분을 차지한다. 본 연구에서는 높은 중의성으로 인하여 오류가 많은 품사를 대상으로, 각 품사에 적합한 자질을 이용하여 학습한, 정확률이 높은 분류기를 통계적 방식의 태거와 순차 결합하여 형태소분석/품사부착 성능을 향상하였다. 2003년 세종계획 품사 부착 말뭉치 200만 어절에서 학습하여 평가를 한 결과 기존 통계적 품사 부착기에 비해 정확도는 0.62% 향상되었으며, 오류는 13.12% 감소하였다.

• The KIPS Transactions:PartB
• /
• v.15B no.1
• /
• pp.17-24
• /
• 2008

This paper proposes Sasang constitution classification through speech signals analysis values and comparison. For this, this paper wishes to propose Taeumin classification method of output values signals that comes out speech signal analysis to connect with process classification of Soeumin through skin diagnosis by first step in the whole system configuration to provide for objective index of Sasang constitution. First of all, these characteristic of voices wish to extract phonetic elements that each Sasang constitution groups' clear features. Also, we wish to classify Taeumin through constitution groups' difference and similarity on the basis of results value. Finally, the effectiveness of this method is verified through the experiments.

• MALSORI
• /
• v.68
• /
• pp.115-126
• /
• 2008

In this paper, we propose a vocabulary coverage improvement method for embedded continuous speech recognition (CSR) using knowledgebase. A vocabulary in CSR is normally derived from a word frequency list. Therefore, the vocabulary coverage is dependent on a corpus. In the previous research, we presented an improved way of vocabulary generation using part-of-speech (POS) tagged corpus. We analyzed all words paired with 101 among 152 POS tags and decided on a set of words which have to be included in vocabularies of any size. However, for the other 51 POS tags (e.g. nouns, verbs), the vocabulary inclusion of words paired with such POS tags are still based on word frequency counted on a corpus. In this paper, we propose a corpus independent word inclusion method for noun-, verb-, and named entity(NE)-related POS tags using knowledgebase. For noun-related POS tags, we generate synonym groups and analyze their relative importance using Google search. Then, we categorize verbs by lemma and analyze relative importance of each lemma from a pre-analyzed statistic for verbs. We determine the inclusion order of NEs through Google search. The proposed method shows better coverage for the test short message service (SMS) text corpus.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.7
• /
• pp.1307-1311
• /
• 2006

In this paper, we propose new speech feature parameter using the Matrix Factorization for appearance part-based features of speech spectrum. The proposed parameter represents effective dimensional reduced data from multi-dimensional feature data through matrix factorization procedure under all of the matrix elements are the non-negative constraint. Reduced feature data presents p art-based features of input data. We verify about usefulness of NMF(Non-Negative Matrix Factorization) algorithm for speech feature extraction applying feature parameter that is got using NMF in Mel-scaled filter bank output. According to recognition experiment results, we confirm that proposed feature parameter is superior to MFCC(Mel-Frequency Cepstral Coefficient) in recognition performance that is used generally.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.49 no.6
• /
• pp.330-333
• /
• 2000

In this paper, a new pitch estimation method is proposed using the Robinson cochlear model. This estimation method is useful in noisy environments and especially very efficient under cochannel in which two speaker voices exist at the same time. For the one speaker speech, the pitch can be extracted from just the neurogram of the Robinson cochlear model. In this case, as the estimation is performed in time domain, the exact pitch period can be detected though the pitch period is various. But in noisy and cochannel cases, the neurogram has many spurious peaks, so we use the autocorrelators in the neurogram to manifest the period. It the autocorrelators are used for the all delays, the large amount of calculations is necessary. Due to this defect, we propose that the autocorrelators are used for the part of the delays on which energy is concentrated. First of all, the proposed algorithm is applied to the one speaker speech, and later to the cochannel speech. And then the results are compared with the autocorrelation pitch detection method.

• The KIPS Transactions:PartB
• /
• v.9B no.2
• /
• pp.231-236
• /
• 2002

The classification of the speech-sound block comes from the multi-resolution analysis property of the discrete wavelet transform, which is used to reduce the computational time for the pre-processing of speech recognition. The merging algorithm is proposed to extract vapid speech-sounds in terms of position and frequency range. It performs unvoiced/voiced classification and denoising. Since the merging algorithm can decide the processing parameters relating to voices only and is independent of system noises, it is useful for extracting valid speech-sounds. The merging algorithm has an adaptive feature for arbitrary system noises and an excellent denoising signal-to-noise ratio and a useful system tuning for the system implementation.