대어휘 연속음성인식을 위한 서브네트워크 기반의 1-패스 세미다이나믹 네트워크 디코딩

정민화,안동훈,Chung Minhwa,Ahn Dong-Hoon 대한음성학회 2004 말소리 Vol.50 No.-

In this paper, we present a one-pass semi-dynamic network decoding framework that inherits both advantages of fast decoding speed from static network decoders and memory efficiency from dynamic network decoders. Our method is based on the novel language model network representation that is essentially of finite state machine (FSM). The static network derived from the language model network [1][2] is partitioned into smaller subnetworks which are static by nature or self-structured. The whole network is dynamically managed so that those subnetworks required for decoding are cached in memory. The network is near-minimized by applying the tail-sharing algorithm. Our decoder is evaluated on the 25k-word Korean broadcast news transcription task. In case of the search network itself, the network is reduced by 73.4% from the tail-sharing algorithm. Compared with the equivalent static network decoder, the semi-dynamic network decoder has increased at most 6% in decoding time while it can be flexibly adapted to the various memory configurations, giving the minimal usage of 37.6% of the complete network size.

한국어 연속음성인식 시스템 구현을 위한 형태소 단위의 발음 변화 모델링

정민화,이경님,Chung Minhwa,Lee Kyong-Nim 대한음성학회 2004 말소리 Vol.49 No.-

In this paper, we describe a cross-morpheme pronunciation variation model which is especially useful for constructing morpheme-based pronunciation lexicon to improve the performance of a Korean LVCSR. There are a lot of pronunciation variations occurring at morpheme boundaries in continuous speech. Since phonemic context together with morphological category and morpheme boundary information affect Korean pronunciation variations, we have distinguished phonological rules that can be applied to phonemes in within-morpheme and cross-morpheme. The results of 33K-morpheme Korean CSR experiments show that an absolute reduction of 1.45% in WER from the baseline performance of 18.42% WER was achieved by modeling proposed pronunciation variations with a possible multiple context-dependent pronunciation lexicon.

N-gram 기반의 유사도를 이용한 대화체 연속 음성 언어 모델링

박영희,정민화,Park Young-Hee,Chung Minhwa 대한음성학회 2003 말소리 Vol.46 No.-

This paper presents our language model adaptation for Korean spontaneous speech recognition. Korean spontaneous speech is observed various characteristics of content and style such as filled pauses, word omission, and contraction as compared with the written text corpus. Our approaches focus on improving the estimation of domain-dependent n-gram models by relevance weighting out-of-domain text data, where style is represented by n-gram based tf/sup */idf similarity. In addition to relevance weighting, we use disfluencies as Predictor to the neighboring words. The best result reduces 9.7% word error rate relatively and shows that n-gram based relevance weighting reflects style difference greatly and disfluencies are good predictor also.

한국어 대어휘 연속음성 인식용 발음사전 자동 생성 및 최적화

이경님,정민화,Lee Kyong-Nim,Chung Minhwa 대한음성학회 2005 말소리 Vol.55 No.-

In this paper, we describe a morpheme-based pronunciation lexicon useful for Korean LVCSR. The phonemic-context-dependent multiple pronunciation lexicon improves the recognition accuracy when cross-morpheme pronunciation variations are distinguished from within-morpheme pronunciation variations. Since adding all possible pronunciation variants to the lexicon increases the lexicon size and confusability between lexical entries, we have developed a lexicon pruning scheme for optimal selection of pronunciation variants to improve the performance of Korean LVCSR. By building a proposed pronunciation lexicon, an absolute reduction of $0.56\%$ in WER from the baseline performance of $27.39\%$ WER is achieved by cross-morpheme pronunciation variations model with a phonemic-context-dependent multiple pronunciation lexicon. On the best performance, an additional reduction of the lexicon size by $5.36\%$ is achieved from the same lexical entries.

기능어용 음소 모델을 적용한 한국어 연속음성 인식

명주현(JooHyun Myung),정민화(Minhwa Chung) 한국정보과학회 2000 한국정보과학회 학술발표논문집 Vol.27 No.1B

의사형태소를 디코딩 단위로 하는 한국어 연속 음성 인식에서는 조사, 어미, 접사 및 짧은 용언의 어간등의 단어가 상당수의 인식 오류를 발생시킨다. 이러한 단어들은 발화 지속시간이 매우 짧고 생략이 빈번하며 결합되는 다른 형태소의 형태에 따라서 매우 심한 발음상의 변이를 보인다. 본 논문에서는 이러한 단어들을 한국어 기능어라 정의하고 실제 의사 형태소 단위의 인식 실험을 통하여 기능어 집합1, 2를 규정하였다. 그리고 한국어 기능어에 기능어용 음소를 독립적으로 적용하는 방법을 제안했다. 또한 기능어용 음소가 분리되어 생기는 음향학적 변이들을 처리하기 위해 Gaussian Mixture 수를 증가시켜 보다 견고한 학습을 수행했고, 기능어들의 음향 모델 스코어가 높아짐에 따른 인식에서의 삽입 오류 증가를 낮추기 위해 언어 모델에 fixed penalty를 부여하였다. 기능어 집합1에 대한 음소 모델을 적용한 경우 전체 문장 인식률은 0.8% 향상되었고 기능어 집합2에 대한 기능어 음소 모델을 적용하였을 때 전체 문장 인식률은 1.4% 증가하였다. 위의 실험 결과를 통하여 한국어 기능어에 대해 새로운 음소를 적용하여 독립적으로 학습하여 인식을 수행하는 것이 효과적임을 확인하였다.

음성 대화 시스템 구성을 위한 의도 예측을 이용한 구어체 대화 인식

강현석(Hyun-Seok Kang),정민화(Minhwa Chung) 한국정보과학회 1998 한국정보과학회 학술발표논문집 Vol.1998 No.2

조음자질을 이용한 한국인 학습자의 영어 발화 자동 발음 평가

류혁수(Ryu, Hyuksu),정민화(Chung, Minhwa) 한국음성학회 2016 말소리와 음성과학 Vol.8 No.4

This paper aims to propose articulatory features as novel predictors for automatic pronunciation assessment of English produced by Korean learners. Based on the distinctive feature theory, where phonemes are represented as a set of articulatory/phonetic properties, we propose articulatory Goodness-Of-Pronunciation(aGOP) features in terms of the corresponding articulatory attributes, such as nasal, sonorant, anterior, etc. An English speech corpus spoken by Korean learners is used in the assessment modeling. In our system, learners’ speech is forced aligned and recognized by using the acoustic and pronunciation models derived from the WSJ corpus (native North American speech) and the CMU pronouncing dictionary, respectively. In order to compute aGOP features, articulatory models are trained for the corresponding articulatory attributes. In addition to the proposed features, various features which are divided into four categories such as RATE, SEGMENT, SILENCE, and GOP are applied as a baseline. In order to enhance the assessment modeling performance and investigate the weights of the salient features, relevant features are extracted by using Best Subset Selection(BSS). The results show that the proposed model using aGOP features outperform the baseline. In addition, analysis of relevant features extracted by BSS reveals that the selected aGOP features represent the salient variations of Korean learners of English. The results are expected to be effective for automatic pronunciation error detection, as well.

음성인식 기반 응급상황관제

이규환(Lee, Kyuwhan),정지오(Chung, Jio),신대진(Shin, Daejin),정민화(Chung, Minhwa),강경희(Kang, Kyunghee),장윤희(Jang, Yunhee),장경호(Jang, Kyungho) 한국음성학회 2016 말소리와 음성과학 Vol.8 No.2

In emergency dispatching at 119 Command & Dispatch Center, some inconsistencies between the ‘standard emergency aid system’ and ‘dispatch protocol,’ which are both mandatory to follow, cause inefficiency in the dispatcher’s performance. If an emergency dispatch system uses automatic speech recognition (ASR) to process the dispatcher’s protocol speech during the case registration, it instantly extracts and provides the required information specified in the "standard emergency aid system,’ making the rescue command more efficient. For this purpose, we have developed a Korean large vocabulary continuous speech recognition system for 400,000 words to be used for the emergency dispatch system. The 400,000 words include vocabulary from news, SNS, blogs and emergency rescue domains. Acoustic model is constructed by using 1,300 hours of telephone call (8 kHz) speech, whereas language model is constructed by using 13 GB text corpus. From the transcribed corpus of 6,600 real telephone calls, call logs with emergency rescue command class and identified major symptom are extracted in connection with the rescue activity log and National Emergency Department Information System (NEDIS). ASR is applied to emergency dispatcher’s repetition utterances about the patient information. Based on the Levenshtein distance between the ASR result and the template information, the emergency patient information is extracted. Experimental results show that 9.15% Word Error Rate of the speech recognition performance and 95.8% of emergency response detection performance are obtained for the emergency dispatch system.

비원어민 한국어 말하기 숙련도 평가와 평가항목의 상관관계

양승희(Yang, Seung Hee),정민화(Chung, Minhwa) 한국음성학회 2017 말소리와 음성과학 Vol.9 No.3

Much research attention has been directed to identify how native speakers perceive non-native speakers’ oral proficiency. To investigate the generalizability of previous findings, this study examined segmental, phonological, accentual, and temporal correlates of native speakers’ evaluation of L2 Korean proficiency produced by learners with various levels and nationalities. Our experiment results show that proficiency ratings by native speakers significantly correlate not only with rate of speech, but also with the segmental accuracies. The influence of segmental errors has the highest correlation with the proficiency of L2 Korean speech. We further verified this finding within substitution, deletion, insertion error rates. Although phonological accuracy was expected to be highly correlated with the proficiency score, it was the least influential measure. Another new finding in this study is that the role of pitch and accent has been underemphasized so far in the non-native Korean speech perception studies. This work will serve as the groundwork for the development of automatic assessment module in Korean CAPT system.

음운변화 규칙을 이용한 음성사전 생성

위선희(Sunhee Wee),정민화(Minhwa Chung) 한국정보과학회 1997 한국정보과학회 학술발표논문집 Vol.1997 No.2