• Title/Summary/Keyword: Pronunciation modeling

Search Result 25, Processing Time 0.02 seconds

Pronunciation Variation Modeling for Korean Point-of-Interest Data Usins Prosodic Information (운율 정보를 이용한 한국어 위치 정보 데이터의 발음 모델링)

  • Kim, Sun-Hee;Park, Jeon-Gue;Jeon, Je-Hun;Na, Min-Soo;Chung, Min-Hwa
    • Annual Conference on Human and Language Technology
    • /
    • 2006.10e
    • /
    • pp.51-56
    • /
    • 2006
  • 일반적으로 운율 정보를 음성인식에 이용한 연구들에 있어서는 대부분 운율의 음향적 정보를 이용하는데 반하여, 본 연구에서는 운율어나 음절수와 같은 운율의 구조적 정보가 인식률 향상에 기여함을 보인다. 본 논문은 두 가지 운율 정보, 즉 운율어와 음절수를 이용하여 발음모델링을 할 경우에 음성인식기의 성능을 평가하는 것을 목표로 하는 것으로, 먼저, 운율어를 이용하여 위치 정보데이터의 가능한 모든 발음을 생성하고, 다시 음절 수를 기준으로 발음변이 수를 조절하는 방법을 제시한 다음, 제안한 방법에 의하여 생성한 발음사전을 이용하여 음성인식의 성능을 평가하였다. 실험결과 운율어를 이용하여 발음 사전을 제작한 모든 경우에 베이스라인과 비교하여 성능이 향상됨을 보였는데, 베이스라인의 WER 4.63% 에서 최대 8.4%의 WER 가 감소하였다. 위치 정보 데이터의 음절수에 따라서 발음 변이의 수를 조절한 결과도 전체적으로는 3 음절로 그 수를 제한한 경우, 6 음절이상 단어에서는 4음절로 제한한 경우에 가장 좋은 인식 성능을 얻을 수 있어서, 음절수에 따른 발음변이 수의 조절이 효과적임을 알 수 있었다.

  • PDF

Robust Speech Recognition Algorithm of Voice Activated Powered Wheelchair for Severely Disabled Person (중증 장애우용 음성구동 휠체어를 위한 강인한 음성인식 알고리즘)

  • Suk, Soo-Young;Chung, Hyun-Yeol
    • The Journal of the Acoustical Society of Korea
    • /
    • v.26 no.6
    • /
    • pp.250-258
    • /
    • 2007
  • Current speech recognition technology s achieved high performance with the development of hardware devices, however it is insufficient for some applications where high reliability is required, such as voice control of powered wheelchairs for disabled persons. For the system which aims to operate powered wheelchairs safely by voice in real environment, we need to consider that non-voice commands such as user s coughing, breathing, and spark-like mechanical noise should be rejected and the wheelchair system need to recognize the speech commands affected by disability, which contains specific pronunciation speed and frequency. In this paper, we propose non-voice rejection method to perform voice/non-voice classification using both YIN based fundamental frequency(F0) extraction and reliability in preprocessing. We adopted a multi-template dictionary and acoustic modeling based speaker adaptation to cope with the pronunciation variation of inarticulately uttered speech. From the recognition tests conducted with the data collected in real environment, proposed YIN based fundamental extraction showed recall-precision rate of 95.1% better than that of 62% by cepstrum based method. Recognition test by a new system applied with multi-template dictionary and MAP adaptation also showed much higher accuracy of 99.5% than that of 78.6% by baseline system.

3D Character Production for Dialog Syntax-based Educational Contents Authoring System (대화구문기반 교육용 콘텐츠 저작 시스템을 위한 3D 캐릭터 제작)

  • Kim, Nam-Jae;Ryu, Seuc-Ho;Kyung, Byung-Pyo;Lee, Dong-Yeol;Lee, Wan-Bok
    • Journal of the Korea Convergence Society
    • /
    • v.1 no.1
    • /
    • pp.69-75
    • /
    • 2010
  • The importance of a using the visual media in English education has been increased. By an importance of Characters in English language content, the more effort is needed for a learner to show the English pronunciation and a realistic implementation. In this paper, we tried to review the Syntax-based Educational Contents Authoring System. For the more realistic lip-sync character, 3D character to enhance the efficiency of the education was constructed. We used a chart of the association structure analysis of mouth's shape. we produced an optimized 3D character through a process of a concept, a modeling, a mapping and an animating design. For more effective educational content for 3D character creation, the next research will be continuously a 3d Character added to a hand motion and body motion in order to show an effective communication example.

A Lingual Sound Analysis based on Oriental Medicine Auscultation for Heart Diseases Diagnosis (심장(心臟) 질환(疾患) 진단(診斷)을 위한 한의학적 청진(聽診) 기반의 설음(舌音) 분석)

  • Kim, Bong-Hyun;Cho, Dong-Uk;Her, Sung-Ho
    • The Journal of Korean Institute of Communications and Information Sciences
    • /
    • v.34 no.8B
    • /
    • pp.830-838
    • /
    • 2009
  • Oriental medicine lacks diagnosis data in fixed quantity possible to express visually to patients by depending on clinician's intuition than Western medicine that continues to development by various diagnosis devices. For that, this paper intends to examine relation between heart and voice signal regarded as center organ and source of life and mind in order to implement objectification through the visualization of oriental diagnosis method above all. According to because the heart is related to the tongue among five organs, by thinking with sounds, we would design the way of identifying existence of heart diseases focused on the fact that lingual sound pronunciation of heart patient is inexact. For this, we achieved a comparison, analysis of statistical bandwidth and morphological modeling of the second formants frequency about a lingual sound for their voice constituted subject group of heart diseases and normal people. Finally, we analyzed interrelationship to the result of experiment by designed method.

Studies on the Construction and the Artificial Mountain Theory of Amisan in the Gyeongbok Palace (경복궁 아미산의 조영과 조산설(造山說)에 관한 고찰)

  • Jung, Woo-Jin;Sim, Woo-Kyung
    • Journal of the Korean Institute of Traditional Landscape Architecture
    • /
    • v.30 no.2
    • /
    • pp.72-89
    • /
    • 2012
  • This study aimed to reconsider the theory that the renowned Amisan(峨眉山) terraced garden at north of Gyotaejeon(交泰殿) was artificially made, by reviewing the historical records and drawings. It has been widely accepted that Amisan was made of the digged soil from Gyeonghoeji(慶會池). But several arguments about artificial mountain theory of Amisan that completely not be found in historical records have been raised in this study. The results were summarized as follows; the inherent contradiction in existing opinion, the discordance between the time of building Gyeonghoeji and Gyotaejeon, the existence of the mountain range which connect Baekaksan and Amisan appeared in Dohyeong(圖形), historical documents written in the years of kingdoms of Youngjo(英祖) and Gojong(高宗), a high position seen from Heungbogjeon(興復殿) in the north Amisan through the wall in the east but impassable, an opinion about realization Amisan as geomantic term of Amisa(蛾眉砂) at the time of Gyeongbok Palace reconstruction, and preservation of the mountain range in Gyeongbok Palace that comes from the result of the arguments in main mountain of Gyeongbok Palace in the year of Sejong(世宗). In addition, it was investigated why the slop in the north of Gyotaejeon was named as Aminsan and why the artificial mountain theory is appeared and made a conclusion that the Amisan comes from the change of the pronunciation of the geomantic term "Amisa", and modeling the yijing[意景] of Amisan which is a sacred place of Taoism and Buddhism in Sichuan[四川] of Chinaand the view of construction to mean defeating a spirit of smallpox which had to be cured. And it seems to be a result which retroactively applied the artificial mountain theory of Amisanis the technique of 'constructing mountain with digged pond dirt' to the relationship between Gyeonghoeji and Amisan. The greater part of mountain range which was connecting with Baekaksan and Amisan was seriously disconnected with large scale of exposition by the Japanese colonial period in 1915. But low slope is kept about 70 meters along the trail northeast of Gyotaejeon. Accordingly, it is judged that the range has not been entirely destroyed. And according to the result of elevation analysis, discontinuous slope form certain axis is found, so the mountain range of Amisan is approximately estimated. This basic research about the mountain range of Amisan might provide a critical clue about restoration of topography in Gyeongbok Palace.