• Journal of Multimedia Information System
• /
• 제5권3호
• /
• pp.209-214
• /
• 2018

Today is the era of artificial intelligence. With the development of artificial intelligence, machines have begun to impersonate various human characteristics today. Chatbot is one instance of this interactive artificial intelligence. Chatbot is a computer program that enables to conduct natural conversations with people. As mentioned above, Chatbot conducted conversations in text, but Chatbot, in this study evolves to perform commands based on speech-recognition. In order for Chatbot to perfectly emulate a human dialogue, it is necessary to analyze the sentence correctly and extract appropriate response. To accomplish this, the sentence is classified into three types: objects, actions, and preferences. This study shows how objects is analyzed and processed, and also demonstrates the possibility of evolving from an elementary model to an advanced intelligent system. By this study, it will be evaluated that speech-recognition based Chatbot have improved order-processing time efficiency compared to text based Chatbot. Once this study is done, speech-recognition based Chatbot have the potential to automate customer service and reduce human effort.

• Journal of Genetic Medicine
• /
• 제12권1호
• /
• pp.19-24
• /
• 2015

Speech and language are uniquely human-specific traits, which contributed to humans becoming the predominant species on earth. Disruptions in the human speech and language function may result in diverse disorders. These include stuttering, aphasia, articulation disorder, spasmodic dysphonia, verbal dyspraxia, dyslexia and specific language impairment. Among these disorders, stuttering is the most common speech disorder characterized by disruptions in the normal flow of speech. Twin, adoption, and family studies have suggested that genetic factors are involved in susceptibility to stuttering. For several decades, multiple genetic studies including linkage analysis were performed to connect causative gene to stuttering, and several genetic studies have revealed the association of specific gene mutation with stuttering. One notable genetic discovery came from the genetic studies in the consanguineous Pakistani families. These studies suggested that mutations in the lysosomal enzyme-targeting pathway genes (GNPTAB, GNPTG and NAPGA) are associated with non-syndromic persistent stuttering. Although these studies have revealed some clues in understanding the genetic causes of stuttering, only a small fraction of patients are affected by these genes. In this study, we summarize recent advances and future challenges in an effort to understand genetic causes underlying stuttering.

• 말소리와 음성과학
• /
• 제9권3호
• /
• pp.75-83
• /
• 2017

The purpose of this study is to find clues to the risk of voice disorders in soprano students. The subjects of the study were 17 soprano students and 18 general students (women). The phonation of vowels /a/, /i/, and /u/ with C4 and F4 notes in each group were recorded. Then, only soprano students were made to record their classical vocalization containing vibrato. Formant, formant energy, bandwidth, VAI (vowel area index), VSA (vowel space area) and L/H ratio were analyzed. There was significant difference in F3 such that the singers' note was measured around 3 kHz which seems to be 400 Hz higher than one from general students. But, There was no significant difference in L/H ratio between soprano student and the general student. There was a significant difference in F3 in the comparison of the soprano students' two vocalization methods. Classical vocalization was measured at 200Hz higher than sustained phonation in F3. Vocal tract adjustment was made and vowel space changed, but there was no significant difference in F3 energy, which is the index of singers' formant according to the phonation method. The L/H ratio, which can be a direct indicator of vocal effort, has no difference in phonation method and is lowered in all phonation methods as the pitch increases. C4 and F4 pitches are lower than the singing range of the soprano. When the pitch changes, vocal effort increases like a general student which will be an indicator of the risk of vocalization. This will be a clue to the vocalization of the immature soprano student.

• 음성과학
• /
• 제8권2호
• /
• pp.107-117
• /
• 2001

The purpose of this study is to revisit the property of tapping, based on the data from Yonbyon Korean. Taps have been described as short segments derived from corresponding stops or trills. It is also widely assumed that tapping occurs due to lenition to minimize articulatory effort. However, Yonbyon Korean data show that taps can occur in strong as well as weak positions The results of the acoustic experiments conducted in this study show that in syllable-onset position, obstruent taps consistently appear from the underlying laterals, while in intervocalic position, sonorant taps similar to American English taps occur. The results of this study provide evidence against the uniform account of tapping as the result of lenition.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2007년도 한국음성과학회 공동학술대회 발표논문집
• /
• pp.311-314
• /
• 2007

Spoken dialog system development includes many laborious and inefficient tasks. Since there are many components such as speech recognizer, language understanding, dialog management and knowledge management in a spoken dialog system, a developer should take an effort to edit corpus and train each model separately. To reduce a cost for editting corpus and training each models, we need more systematic and efficent working environment. For the working environment, we propose DialogStudio as an spoken dialog system workbench.

• 말소리와 음성과학
• /
• 제8권3호
• /
• pp.39-49
• /
• 2016

The present study examined the change of aerodynamic features after laryngomicrosurgery in patients with vocal polyps. Aerodynamic evaluation was performed in thirty-nine patients (15 males and 24 females) one week before surgery and four weeks after surgery. Evaluation protocols of vital capacity, maximum sustained phonation(MXPH), and voicing efficiency(VOFT) were used to collect 29 phonatory aerodynamic measures, requiring voice with a comfortable pitch and loudness. Statistically significant changes were found for phonation time and airflow values in the MXPH protocol, while changes were also found for airflow values, subglottal pressure values and acoustic resistance values in the VOFT protocol. Although phonation time was increased in both male and female patients, gender-dependent changes were found in airflow measurements. Men's phonation time increased with no difference in airflow rate, but women's phonation time increased with decreased airflow rate and lower subglottal pressure. The changes of aerodynamic features may be affected by women's self-perceived change for vocal attitude, which was reducing sense of vocal effort after surgery.

• 말소리와 음성과학
• /
• 제2권2호
• /
• pp.93-100
• /
• 2010

The purpose of this study was to investigate the differences of feeding ability and articulatory function in the children with spastic cerebral palsy and typically developing children according to feeding subtypes. The feeding subtypes were limited by chewing, cup drinking and spoon feeding. 14 children with spastic cerebral palsy and 14 typically developing children were participated in this study. The results were following as; First, there were significant differences in overall articulatory function between two groups Second, all scores of articulators' function according to feeding subtypes in children with cerebral palsy was significantly higher than typically developing children Third, chewing mode in feeding subtypes was highly correlated with lip and tongue movement. compared to another Finally, the correlation between spoon feeding and mobility of lip and tongue was high in both groups. These results suggested that These results suggest that the effort to find out the differences feeding ability and appliances for articulatory function in CP children are meaningful in catching their speech ability indirectly. Moreover, the more organized feeding skills should be discussed in the relationship with verbal and nonverbal development.

• 한국음향학회지
• /
• 제25권5호
• /
• pp.213-221
• /
• 2006

합성음의 음질을 향상시키기 위하여 분할된 corpora로부터 합성유닛을 선택하여 사용하는 연속음성합성에서 정확한 음소분할은 매우 중요하다. 일반적으로 음소분할은 사람에 의해 수행되지만 많은 작업량으로 인한 시간적 지연, 일관 성 유지 어려움 등 많은 문제가 발생한다. 이에 따라 음성인식에서 도입된 HMM 기반의 자동음소분할이 음성인식, 음성 합성에서 널리 사용되어지고 있지만 음성전문가의 수작업 결과와 비교할 때 HMM 기반 자동음소분할은 오류가 있고, 이는 합성음 품질의 열화의 주요 원인이 되고 있다. 본 논문에서는 HMM 기반의 자동음소분할기를 사용하여 나타난 자동음소분할 결과와 수작업에 의한 음소분할 결과를 비교하고 유형별로 분석함으로써 음성합성의 성능향상을 위해 개선해야 할 문제점들을 제시한다. 실험에서는 ETRI의 표준형 한국어 공통 음성 DB을 사용하였고, 오차의 범위가 20ms를 벗어난 경우를 분절 오류로 간주하였다. 실험 결과 여성화자의 경우 파열음 + 모음, 파찰음 + 모음, 모음 + 유음 음소쌍에서는 각각 약 99%, 99.5%, 99%의 높은 정확률을 보인 반면, 폐쇄음 + 비음, 폐쇄음 + 유음, 비음 + 유음 음소쌍에서는 44.89%, 50%, 55% 의 낮은 정확률을 보였으며, 남성화자에 대한 실험결과에서도 유사한 경향을 보였다.

• 한국정보과학회논문지:정보통신
• /
• 제28권4호
• /
• pp.540-549
• /
• 2001

인터넷전화 서비스는 저렴한 가격과, 타 서비스와 통합 및 가치부가(Value Added)면에서 기존의 전화에 비해 많은 장점을 가지고 있으나, 상대적으로 낮은 음질로 인하여 사용자의 요구를 만족시키지 못하고 있다. 이것은 현재 인터넷은 best-effort형 패킷 전달 서비스만을 제공하고 있기 때문에 전송지연, 패킷손실, 지터 등을 보장할 수 있는 방법이 없기 때문이다. 본 논문에서는 인터넷전화에서 패킷손실이나 전송지연으로 인한 음질 저하문제를 SOLA 알고리즘을 이용해 보완하였다. SOLA 알고리즘은 시간축 변환(Time Scaled Modification) 기법중의 하나로써 음성신호가 가지는 중요한 스펙트럼 정보는 그대로 유지하면서 단지 발음 속도만을 변환시키는 기법이다. 본 논문에서는 송신측에서 패킷을 전송하면 수신측에서는 수신 패킷에 SOLA 알고리즘을 적용하여 수신 패킷을 사람이 인지하지 못하는 수준에서 확장하여 전송지연으로 인한 패킷손실을 감소시킨다. 시뮬레이션 결과 전송지연으로 인한 패킷 손실 확률이 상당히 감소되었고 음질 또한 상당히 개선되었다.

• 한국전자통신학회논문지
• /
• 제9권11호
• /
• pp.1221-1226
• /
• 2014

본 논문에서는 음성 인식의 효율을 높이기 위하여 음소 클러스터 개수의 효과에 대해 연구하였다. 이를 위하여 음소 클러스터 개수를 바꾸어 가면서 수정된 k-평균 군집 알고리듬을 사용하여 코우드북을 작성하였다. 그런 다음, 퍼지 벡터 양자화와 은닉 마코브 모델을 사용하여 음성인식 테스트를 수행하였다. 실험 결과 두 개의 영역이 구분되어 나타났다. 음소 클러스터 개수가 클 때 인식 성능은 대체로 그와 무관하지만, 개수가 작을 때에는 그 감소와 더불어 인식 오류율이 비선형적으로 증가하는 것으로 나타났다. 수치 해석적 계산으로부터, 이 비선형 영역은 멱승함수에 의해 모델링 될 수 있었다. 또한 300개의 고립단어 인식의 경우에, 166개의 음소클러스터가 최적의 수임을 보일 수 있었다. 이는 음소당 3개 정도의 변화에 해당하는 값이다.