• Proceedings of the Korea Institutes of Information Security and Cryptology Conference
• /
• 2003.07a
• /
• pp.204-207
• /
• 2003

생체인식 기술은 전통적인 비밀번호 방식 또는 토큰 방식보다 신뢰성 면에서 더 선호되지만, 환경의 영향에 매우 민감하여 성능의 한계가 있다. 이러한 단일 생체인식 기술의 한계를 극복하기 위하여 여러 종류의 생체 정보를 결합한 다중 생체인식 (multimodal biometrics)에 관한 다양한 연구가 진행되고 있다 본 논문에서는 다중 생체인식 기술을 간략히 소개하고, Support Vector Machines(SVM)을 이용하여 얼굴 및 음성 정보를 함께 이용한 다중 생체인식 실험으로 성능이 개선될 수 있음을 확인하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10a
• /
• pp.811-813
• /
• 2004

휴대전화에서 전달되는 음성데이터들이 전달되는 과정에서 잡음 등의 외부 요인으로 인하여 데이터에 손실이 생기는 문제가 발생한다. 이렇게 전달된 음성데이터가 음성 인식기를 통과하면 바로 음성 인식기를 통과했을 때 보다 인식률이 낮아진다. 본 연구에서는 음성인식 알고리즘을 이용하여 홈 네트워크를 제어하는데 있어서 음성 인식율을 향상시키기 위해서 반복적으로 음성 데이터를 입력받아. 이를 유사율 알고리즘을 적용시켜 추출 된 여러 개의 데이터(text)를 이미 구축된 홈 네트워크 용어 관련 사전에 등록된 단어와의 유사성을 검토하여 추출된 결과로 홈 네트워크를 제어하는 방안을 제안한다. 이 결과, 기존의 방법에 비해서 10% 정도의 인식률의 향상을 확인할 수 있었다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.10 no.3
• /
• pp.319-325
• /
• 2015

Remarkable advances in wireless communication technology have enabled communications among people who are far away from each other. In recent, the needs of local area voice communication using a wireless system based on low-cost and simple hardware are rapidly rising. However, since these applications require that the multi-users communicate on the same wireless channel in a small area, the existing voice technologies are not suitable for directly applying to these applications. Therefore, in this paper I propose a novel idea enabling multi-user voice communication. In particular, as a short range wireless solution, I employ the IEEE 802.15.4 based on low power and low cost. However, since originally the standard is not developed for voice communication, we extend the original scheme to be suitable for the voice communication by utilizing the extended concept of GTS. The capacity and validity of the proposed scheme are evaluated through quantitative analysis in various voice compression rates.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.37 no.1
• /
• pp.32-43
• /
• 2000

In this paper, we investigate the existing medium access control (MAC) protocols to integrate the voice and data services in packet-based CDMA networks and furthermore, propose a new approach to circumvent the operational limits inherent in them. We propose the $P^2R$-CDMA (Prioritized Packet Reservation Code Division Multiple Access) protocol for the uplink in the synchronous multi-code CDMA system, which employs the centralized frame-based slot reservation along with the dynamic slot assignment in the base station using the QoS-oriented dynamic priority of individual terminal. The simulation results show that, as compared with the existing scheme based on the adaptive permission probability control (APC), the proposed approach can significantly improve the system capacity while guaranteeing the real-time requirement of voice service.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.20 no.5
• /
• pp.133-138
• /
• 2010

We present a new multimodal biometric authentication method, which performs both feature-level fusion and decision-level fusion. After generating support vector machines for new features made by integrating face and voice features, the final decision for authentication is made by integrating decisions of face SVM classifier, voice SVM classifier and integrated features SVM clssifier. We justify our proposal by comparing our method with traditional one by experiments with XM2VTS multimodal database. The experiments show that our multilevel fusion algorithm gives higher recognition rate than the existing schemes.

• Proceedings of the IEEK Conference
• /
• 1999.06a
• /
• pp.661-666
• /
• 1999

웨이브렛 이론은 응용수학에서 처음 소개된 후 다중해상도 표면 및 이산신호의 부대역 분해방법 등에 대한 단일화된 이론을 제공하고 있으며 최근 신호처리 전반에 걸쳐 널리 이용되고 있는 이론이다. 본 논문에서는 최근 들어 신호저리분야의 새로운 기법으로 제시된 웨이브렛 이론에 대한 소개와 더불어 이를 이용하여 음성개선, 유성음/무성음/묵음 판별, 끝점검출, 피치 및 성문 폐쇄시점 검출 등의 음성신호처리에 적용한 예들을 소개한다.

• Journal of the Korea Society of Computer and Information
• /
• v.3 no.4
• /
• pp.103-109
• /
• 1998

Domastic telephony transmission networks mainly using El in 2.048Mb㎰ is composed to 30 channels and each channel is assigned to 64Kb㎰ voice coding rate. El method always uses TDM, so it is fixed channels. In this paper, it shows that using CDM enlarge the subscribers and voice channels

• 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.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.5
• /
• pp.390-399
• /
• 2004

In this paper, we design and implement a multiple-view VoiceXML editor to improve editing efficiency of the VoiceXML. The VoiceXML multiple-view Editor uses a MVC framework to support multiple views and paradigm. Our multiple-view editor consists of Model. View and Controller using MVC framework. A model, core data structure. is constructed of abstract syntax tree and abstract grammar. A view. user interface. is formalized in unparsing rules and unparser. A controller. to control model and view. is made of command interpreter and tree handler. The VoiceXML multiple-view editor overcomes a drawbacks of existing XML editors by showing document structure and context concurrently. as well as document flows. Our VoiceXML multiple-view editor. which MVC framework has been applied, provides various editing views concurrently to users. Thereby. it supports efficient and convenient editing environments for voice-web documents to users and it guarantees transparency of editors. as various views have a same consistent model.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.1
• /
• pp.93-101
• /
• 1996

In this paper, we implement the system for speech processing to provide the subscribers who are using the telephone network with better speech services. We develop the specified board which is processing speech signal and devise the system which carries out storing and replaying the speech signal under the condition that one master board controls multiple DSP(Digital Signal Processing) boards using VME bus. We use CPU30 board as a maste board and develop SPM(Signal Processing Module) board as a DSP board and then evaluate performance of the system.