• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.11
• /
• pp.1221-1226
• /
• 2014

In an effort to improve the efficiency of the speech recognition, we investigate the effect of the number of phoneme clusters. For this purpose, codebooks of varied number of phoneme clusters are prepared by modified k-means clustering algorithm. The subsequent processing is fuzzy vector quantization (FVQ) and hidden Markov model (HMM) for speech recognition test. The result shows that there are two distinct regimes. For large number of phoneme clusters, the recognition performance is roughly independent of it. For small number of phoneme clusters, however, the recognition error rate increases nonlinearly as it is decreased. From numerical calculation, it is found that this nonlinear regime might be modeled by a power law function. The result also shows that about 166 phoneme clusters would be the optimal number for recognition of 300 isolated words. This amounts to roughly 3 variations per phoneme.

• KIISE Transactions on Computing Practices
• /
• v.21 no.10
• /
• pp.631-638
• /
• 2015

Today, the demand for speech recognition systems in mobile environments is increasing rapidly. This paper proposes a novel method for Korean phoneme segmentation that is applicable to a phoneme based Korean speech recognition system. First, the input signal constitutes blocks of the same size. The proposed method is based on a volatility indicator calculated for each block of the input speech signal, and the bulk indicators calculated for each bulk in blocks, where a bulk is a set of adjacent samples that have the same sign as that of the primitive indicators for phoneme segmentation. The input signal vowels, voiced consonants, and voiceless consonants are sequentially recognized and the boundaries among phonemes are found using three devoted recognition algorithms that combine the two types of primitive indicators. The experimental results show that the proposed method can markedly reduce the error rate of the existing phoneme segmentation method.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.8
• /
• pp.694-702
• /
• 2003

The word intelligibility test for dysarthric speakers was designed to examine phonetic contrasts that are likely (1) to be sensitive to intelligibility impairment and (2) to contribute significantly to speech intelligibility. These phonetically contrasting word pairs were tested and proved to be reliable and to be valid, The results showed that in Korean dysarthric patients, the percentage of error in final position contrast was higher than in any other position. Unlike the results of previous studies, the initial-position contrasts were crucial in predicting the overall intelligibility among Korean patients.

• Proceedings of the KSLP Conference
• /
• 2003.11a
• /
• pp.241-242
• /
• 2003

연구목적 : 청각장애 아동들은 청각적 피드백의 손실로 인하여 분절적 측면뿐만 아니라 초분절적인 측면도 건청 아동과는 다른 형태를 나타낸다. 석동일(1999)은 청각장애인의 모음 조음의 특성을 고찰한 결과, 저모음의 지속시간이 길며, 고모음의 지속시간이 짧다고 하였다. 또한, 청각장애인들은 자음 산출에 있어서 가시적인 효과가 높은 음소가 낮은 음소에 비해 정조음률이 높다. 따라서, 본 연구의 목적은 청각장애 아동의 자음 중 가장 많이 오조음하는 /s/의 오류 형태에 따라 선.후행하는 고모음 /i/와 저모음 /a/의 지속시간을 비교.연구하는 것이다. (중략)

• Journal of Korea Entertainment Industry Association
• /
• v.14 no.8
• /
• pp.213-219
• /
• 2020

The purpose of this study was to investigate the characteristics of error types in naming task for 8 questionable demeatia groups, 9 definite dementia groups, and 10 normal groups. The items of naming error analysis were classified into visual perception errors, semantic association errors, semantic non-correlation errors, phoneme errors, Don't Know, and No Response. For the analysis, descriptive statistics analysis, analysis of variance, and multivariate analysis of variance were conducted using SPSS 21.0. As a result, there was a significant difference in the error rate between groups according to the error type. The errors that showed significant differences between the normal group and the other two groups were visual perception errors and semantic non-related errors. The error of non-response was different from the dementia confirmation group, but there was no significant difference from the dementia suspicion group. These results showed that Alzheimer's patients had a defect in confrontation naming ability. Also, it was found that it is appropriate to provid other clues when the defects caused by the degeneration of a specific step during the information processing process become severe.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.1
• /
• pp.99-107
• /
• 2022

This study attempted to analyze the consonant perception ability and errors and to verify compatibility items for hearing impaired children wearing hearing aids and cochlear implants using the Phoneme Perception Test for School-Aged children (PPT-S). As a result of the study, it was found that children with hearing impairments have more difficulty in perceiving final consonants than initial consonants. The hard type of PPT-S, in which the articulation method and articulation place of the target and foil words are similar, felt more difficult than the easy type. Among the initial consonants, the incorrect response rate for aspiration sound was higher. In the case of final consonants, the incorrect answer rate for 'ㄷ' and 'ㅁ' was relatively higher. There was no significant difference in the percentage of correct response rate according to the gender of the speaker. The above results can be usefully used as basic data for standardizing of PPT-S and evaluating the intervention effects before and after hearing rehabilitation with hearing impaired children.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.2
• /
• pp.27-34
• /
• 2001

Utterance verification is used in variable vocabulary word recognition to reject the word that does not belong to in-vocabulary word or does not belong to correctly recognized word. Utterance verification is an important technology to design a user-friendly speech recognition system. We propose a new utterance verification algorithm for no-training utterance verification system based on the minimum verification error. First, using PBW (Phonetically Balanced Words) DB (445 words), we create no-training anti-phoneme models which include many PLUs(Phoneme Like Units), so anti-phoneme models have the minimum verification error. Then, for OOV (Out-Of-Vocabulary) rejection, the phoneme-based confidence measure which uses the likelihood between phoneme model (null hypothesis) and anti-phoneme model (alternative hypothesis) is normalized by null hypothesis, so the phoneme-based confidence measure tends to be more robust to OOV rejection. And, the word-based confidence measure which uses the phoneme-based confidence measure has been shown to provide improved detection of near-misses in speech recognition as well as better discrimination between in-vocabularys and OOVs. Using our proposed anti-model and confidence measure, we achieve significant performance improvement; CA (Correctly Accept for In-Vocabulary) is about 89％, and CR (Correctly Reject for OOV) is about 90％, improving about 15-21％ in ERR (Error Reduction Rate).

• Proceedings of the KSPS conference
• /
• 1996.10a
• /
• pp.468-471
• /
• 1996

이 논문에서는 대량의 음성합성용 운율 DB를 용이하게 구축하기 위해 음성번역시스템을 이용한 자동 레이블러의 성능을 다양한 음성데이타를 대상으로 평가하였다. 실험 결과 FM radio news문장, 대화체 문장 및 낭독체 문장 등에는 레이블링 대상 음소의 약 80% 이상이 오류가 30msec 이내인 범위로 레이블링 되며, 고립단어에 대해서는 약 60%의 성능을 보여주고 있다. 현재 당 연구실에서는 자동 레이블러를 이용하여 합성용 운율 DB 및 합성단위를 작성하고 있으며. 자동 레이블러를 이용함으로서 일관성 있는 레이블링 결과를 얻을 수 있을 환 아니라 작성하는데 소요되는 시간도 줄일 수 있었다

• Proceedings of the KSPS conference
• /
• 2005.11a
• /
• pp.219-222
• /
• 2005

본 논문에서는 한국어 TSS 시스템을 위한 언어처리부의 설계 및 구현 과정을 설명한다. 구현된 언어처리부는 형태소 분석, 품사 태깅, 발음 변환 과정을 거쳐, 주어진 문장의 가장 적절한 발음열과 각 음소의 해당 품사를 출력한다. 프로그램은 표준 C언어로 구현되어 있고, Windows와 Linux에서 모두 동작되는 것을 확인하였다. 수동으로 품사가 할당된 4.5만 어절의 코퍼스로부터 형태소 사전을 구축하였으며, 모든 단어가 사전에 등록되어 있다고 가정할 경우, 488문장의 실험 자료에 대해 어절 단위 오류율이 3.25%이었다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.21-24
• /
• 1999

본 논문에서는 가변어휘 음성인식기의 성능개선 작업에 관한 내용을 기술하고 있다. 묵음을 포함한 총 40개의 문맥독립 음소모델을 사용한다. LDA 기법을 이용하여 동일차수의 특징벡터내에 보다 유용한 정보를 포함시키고, likelihood 계산시 가우시안 분포와 mixture weight에 대한 가중치를 달리 함으로써 성능향상을 볼 수 있었다. ETRI POW 3848 DB만을 사용하여 실험한 경우, $21.7\%$의 오류율 감소를 확인할 수 있었다. 잡음환경 및 어휘독립환경을 고려하여 POW 3848 DB와 PC 168 DB 및 PBW445 DB를 사용한 실험도 행하였으며, PBW 445 DB를 사용한 어휘독립 인식실험의 경우 $56.8\%$의 오류율 감소를 얻을 수 있었다.