• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.515-518
• /
• 2004

This paper presents a method that improves the performance of the personal voice dialling system in which speaker Independent phoneme HMM's are used. Since the speaker independent phoneme HMM based voice dialing system uses only the phone transcription of the input sentence, the storage space could be reduced greatly. However, the performance of the system is worse than that of the system which uses the speaker dependent models due to the phone recognition errors generated when the speaker Independent models are used. In order to solve this problem, a new method that jointly estimates transformation vectors for the speaker adaptation and transcriptions from training utterances is presented. The biases and transcriptions are estimated iteratively from the training data of each user with maximum likelihood approach to the stochastic matching using speaker-independent phone models. Experimental result shows that the proposed method is superior to the conventional method which used transcriptions only.

• Proceedings of the KSPS conference
• /
• 2007.05a
• /
• pp.123-126
• /
• 2007

This paper suggests a method to improve the performance of the pathological/normal voice classification. The effectiveness of the mel frequency-based filter bank energies using the fisher discriminant ratio (FDR) is analyzed. And mel frequency cepstrum coefficients (MFCCs) and the feature vectors through the linear discriminant analysis (LDA) transformation of the filter bank energies (FBE) are implemented. This paper shows that the FBE LDA-based GMM is more distinct method for the pathological/normal voice classification than the MFCC-based GMM.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2451-2454
• /
• 2003

In this paper, we propose the computation reduction methods of LSP(Line spectrum pairs) transformation that is mainly used in CELP vocoders. In order to decrease the computational time in real root method the characteristic of four proposed algorithms is as the following. First, scheme to reduce the LSP transformation time uses met scale. Developed the second scheme is the control of searching order by the distribution characteristic of LSP parameters. Third, scheme to reduce the LSP transformation time uses voice characteristics. Developed the fourth scheme is the control of searching interval and order by the distribution characteristic of LSP parameters. As a result of searching time, computational amount, transformed LSP parameters, SNR, MOS test, waveform of synthesized speech, speech, spectrogram analysis, searching time reduced about 37.5％, 46.21％, 46.3％, 51.29％ in average, computational amount is reduced about 44.76％, 49.44％, 47.03％, 57.40％. But the transformed LSP parameters of the proposed methods were the same as those of real root method.

• Proceedings of the KAIS Fall Conference
• /
• 2010.05a
• /
• pp.257-260
• /
• 2010

In this paper, we propose the computation reduction methods of LSP(Line spectrum pairs) transformation that is mainly used in CELP vocoders. In order to decrease the computational time in real root method the characteristic of four proposed algorithms is as the following. First, scheme to reduce the LSP transformation time uses mel scale. Developed the second scheme is the control of searching order by the distribution characteristic of LSP parameters. Third, scheme to reduce the LSP transformation time uses voice characteristics. Developed the fourth scheme is the control of searching interval and order by the distribution characteristic of LSP parameters. As a result of searching time, computational amount, transformed LSP parameters, SNR, MOS test, waveform of synthesized speech, spectrogram analysis, searching time is reduced about 37.5%, 46.21%, 46.3%, 51.29% in average, computational amount is reduced about 44.76%, 49.44%, 47.03%, 57.40%. But the transformed LSP parameters of the proposed methods were the same as those of real root method.

• The Journal of the Korea Contents Association
• /
• v.20 no.3
• /
• pp.432-439
• /
• 2020

This study aims at exploring features of Korean university students' written text, focusing on the written voice, rhetorical patterns, and writing practices through English letters. The data comprised examples of students' English job applications, and a 'purpose-will' model was adopted for the data analysis. The findings showed that the students used unique ways of strategies to convey their voice in a recontextualized setting. Their written voice in the job applications were various, and nobody applied the Korean convention of weather opening. Their rhetorical patterns were a transformation from convergence to divergence, showing integrated patterns of written voice. Students' writing practices revealed their internal values of writing for a task, and they do not directly learn from the teacher's syllabus. This supports the sociocultural framework that learning is a situated activity in a specific discourse community. The study concludes that writing teachers should understand that life-world and learning experience can impact on students' written voice and practices.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.17 no.6
• /
• pp.744-750
• /
• 2014

In this paper, I have measured NCO(Network Centric Operation) Effectiveness of Anti-piracy Operation at the Chunghae Unit. For quantitative analysis, Network Centric Operations Conceptual Framework(U.S Office of Force Transformation) is applied. In accordance with the framework, the Chunghae unit anti-piracy operation scenario is analysed. The scenario is devided with two case(only voice communication and networking). The element of analysis be composed of the organic information, networking, share-ability, and individual information. As a result of analysis, the individual information of first case(only voice) gets 0.59 points. The other side, second case (networking) gets 1 points. This means that NCO has effect on the Chunghae Unit's mission. In addition, I stated the tactics advantage of NCO related a fighting power.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10c
• /
• pp.592-594
• /
• 2003

무선 단말기의 보급과 네트워크 기술의 발전은 무선 단말기를 이용한 인터넷 접속을 증가시키고 있다. 그러나 대부분의 웹 페이지들이 데스크탑에 최적화 되어 있어 무선 단말기를 이용하여 사용자가 원하는 부분에 접근하기까지 반복적인 스크롤링을 해야하는 불편한 점이있다. 기존의 대부분 연구들이 웹페이지를 요약하는 기법을 제안하였지만, 대부분의 웹 페이지들은 한 페이지에 세분화된 섹션과 많은 내용을 담고 있기 때문에 제한된 화면과 입력장치를 가진 무선단말기에 대한 최적화된 해결책이라고 할 수 없다. 이런 문제점을 해결하기 위해 본 논문에서는 웹의 뉴스 페이지내의 뉴스의 섹션을 추출하고. 무선 환경에 적합하도록 VoiceXML형태로 변환해 주는 기법을 제안한다. 본 논문에서 제안된 기법을 통해 사용자는 무선 단말기의 각종 단점을 극복함과 동시에 뉴스에서 선호하는 섹션의 맞춤형 뉴스 서비스를 제공받을 수 있다.

• Proceedings of the KIEE Conference
• /
• 1988.07a
• /
• pp.472-476
• /
• 1988

As to the development of the information society, the common communication network which processes the service for data, image and voice etc., is required. So the higher degree of reliability becomes more important. Therefore, this paper describes the improvement of reliability using the format transformation of the mBIZ code and the reframe usually practiced at the terminal repeater.

• 한국정보통신설비학회:학술대회논문집
• /
• 2008.08a
• /
• pp.584-587
• /
• 2008

In this paper, we propose a real time video codec adapter for enabling video communications with terminals having a codec which is different from each other. When multimedia services are playing with an office service phone such as a video phone or software phone which has video capability, each terminal is not being considered to have optimized video or voice codec. So when a video phone with only one type of video codec is used in the video streaming service which requires another type of codec, the streaming service is not successful without codec transformation. The real time codec adapter in this paper provides a real time code transformation which enables communication services such as video conferencing between terminals which have different codec.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.5
• /
• pp.455-461
• /
• 2002

This paper presents a method that improves the performance of the personal voice dialling system in which speaker independent phoneme HMM's are used. Since the speaker independent phoneme HMM based voice dialing system uses only the phone transcription of the input sentence, the storage space could be reduced greatly. However, the performance of the system is worse than that of the system which uses the speaker dependent models due to the phone recognition errors generated when the speaker independent models are used. In order to solve this problem, a new method that jointly estimates transformation vectors for the speaker adaptation and transcriptions from training utterances is presented. The biases and transcriptions are estimated iteratively from the training data of each user with maximum likelihood approach to the stochastic matching using speaker-independent phone models. Experimental result shows that the proposed method is superior to the conventional method which used transcriptions only.