• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1042-1045
• /
• 2003

In this paper, we propose the speaker identification system that uses vowel that has speaker's characteristic. System is divided to speech feature extraction part and speaker identification part. Speech feature extraction part extracts speaker's feature. Voiced speech has the characteristic that divides speakers. For vowel extraction, formants are used in voiced speech through frequency analysis. Vowel-a that different formants is extracted in text. Pitch, formant, intensity, log area ratio, LP coefficients, cepstral coefficients are used by method to draw characteristic. The cpestral coefficients that show the best performance in speaker identification among several methods are used. Speaker identification part distinguishes speaker using Neural Network. 12 order cepstral coefficients are used learning input data. Neural Network's structure is MLP and learning algorithm is BP (Backpropagation). Hidden nodes and output nodes are incremented. The nodes in the incremental learning neural network are interconnected via weighted links and each node in a layer is generally connected to each node in the succeeding layer leaving the output node to provide output for the network. Though the vowel extract and incremental learning, the proposed system uses low learning data and reduces learning time and improves identification rate.

• Journal of Korea Multimedia Society
• /
• v.6 no.7
• /
• pp.1131-1138
• /
• 2003

Active noise control(ANC) is an approach to noise reduction in which a secondary noise source destructively interferes with the unwanted noise is introduced. Generally, the performance of ANC is determined how well a secondary noise tracks noises. A secondary noise is generated from the cancelling speaker and a error sensor pick up error signal. The transfer function between the cancelling speaker and the error sensor is not flat and distorts secondary noises. Consequently, the performance of ANC is degraded by the transfer function. In this paper, a single sensor ANC which considers the characteristics of the speaker and the error sensor is proposed. To reduce distortion of secondary noises, the transfer function is estimated by adaptive inverse modelling and the primary noises are estimated by Kalman filter. Experimental results show that the proposed single sensor ANC effectively attenuates noises.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.1
• /
• pp.27-31
• /
• 2012

In this paper, we constructed an array of speaker system with directivity by using FXLMS(Filtered-X LMS) algorithm and confirmed its directivity. The front $0^{\circ}$ characteristics of the controlled speaker was suppressed by interfering it with the control signal produced with filter coefficients optimized with respect to the $180^{\circ}$ characteristics of the rear speakers. The directivity of the array of rear speakers was measured and the damping effect of the signal from the front speaker array was confirmed. The frequency characteristics and directivity was investigated by using the adaptive filter coefficients on damping, the damping on the control point was verified in all the frequency range. In 100Hz, 200Hz, 1000Hz regime, the damping effect was observed in the range of front $60^{\circ}{\sim}100^{\circ}$.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.1E
• /
• pp.16-20
• /
• 2005

A novel scheme to measure the speaker information in speech signal is proposed. We develope the theory of quantitative measurement of the speaker characteristics in the information theoretic point of view, and connect it to the classification error rate. Homomorphic analysis based features, such as mel frequency cepstral coefficient (MFCC), linear prediction cepstral coefficient (LPCC), and linear frequency cepstral coefficient (LFCC) are studied to measure speaker specific information contained in those feature sets by computing mutual information. Theories and experimental results provide us quantitative measure of speaker information in speech signal.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.799-800
• /
• 2006

This thesis implemented security systems of voice awareness which is higher accessible than existing security system using biological authentication system and is inexpensive in module of security device, and has an advantage in usability. Proposed the security systems of voice awareness implemented algorithm for characteristic extraction of inputted speaker's voice signal verification, and also implemented database of access control that is founded on extractible output. And a security system of voice awareness has a function of an authority of access control to system.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.154.2-154
• /
• 2001

This paper presented a adaptation approach based on maximum model distance (MMD) method. This method shares the same framework as they are used for training speech recognizers with abundant training data. The MMD method could adapt to all the models with or without adaptation data. If large amount of adaptation data is available, these methods could gradually approximate the speaker-dependent ones. The approach is evaluated through the phoneme recognition task on the TIMIT corpus. On the speaker adaptation experiments, up to 65.55% phoneme error reduction is achieved. The MMD could reduce phoneme error by 16.91% even when ...

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.99.1-99
• /
• 2001

This paper presented an adaptation approach based on maximum model distance (MMD) method. This method shares the same framework as they are used for training speech recognizers with abundant training data. The MMD method could adapt to all the models with or without adaptation data. If large amount of adaptation data is available, these methods could gradually approximate the speaker-dependent ones. The approach is evaluated through the phoneme recognition task on the TIMIT corpus. On the speaker adaptation experiments, up to 65.55% phoneme error reduction is achieved. The MMD could reduce phoneme error by 16.91% even when only one adaptation utterance is used.

• Journal of Satellite, Information and Communications
• /
• v.8 no.4
• /
• pp.17-20
• /
• 2013

In this study, we introduce an LED Driving circuit in order that the information state can indicate audio signal gain and radiate pattern of ultrasonic speaker system for a sign board. Ultrasonic speaker system decreases energy loss and transmits the sound farther. Ultrasonic speaker having such characteristics is useful in that it can be widely used in daily life. Additionally, Proposed LED circuit indicates the information state as linear LED brightness taken from interface of ultrasonic speaker system. Designed circuit is confirmed through $0.35{\mu}m$ CMOS process by Dong-bu.

• Journal of the Korean Society for Precision Engineering
• /
• v.11 no.1
• /
• pp.108-113
• /
• 1994

Active noise control technique has superior performance in low frequency ranges(50 .approx. 400Hz) to the conventional passive noise control technique. For the feasibility of active noise control, it is required to develop a controller which can implement control algorithm on real-time. In this study, therefore, real-time controller is developed using TMS320c25, high speed digital processor. Unlike conventional DSP board of complete ADD ON type, it is possible for the developed controller to interface with the other computer system easily by series communication for the convenience of program development. Furthermore it is designes to be separated readily as a control device. Active noise control of duct system is implemented ti evaluate a performance of developed device. Active noise control of duct system is implemented to evaluate a performance of developed controller using filtered-x LMS algorithm.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2014.10a
• /
• pp.955-956
• /
• 2014

Active noise control is noise reduction method by generate anti-phase control signal for destructive interference of through control speaker. purpose of this paper is create a virtual control source at a using the DBAP(Distance Based Amplitude Panning) algorithm which is one of the three-dimensional sound reproduction method, and verified through the experimentally for noise control method through the virtual control source. We compared active noise method by using one control speaker with active noise control method by using DBAP algorithm.