Go to the main menu
Skip to content
Go to bottom
REFERENCE LINKING PLATFORM OF KOREA S&T JOURNALS
> Journal Vol & Issue
The Journal of the Acoustical Society of Korea
Journal Basic Information
Journal DOI :
The Acoustical Society of Korea
Editor in Chief :
Volume & Issues
Volume 11, Issue 6 - Dec 1992
Volume 11, Issue 4 - Aug 1992
Volume 11, Issue 3 - Jun 1992
Volume 11, Issue 2 - Apr 1992
Volume 11, Issue 1 - Feb 1992
Volume 11, Issue 1E - 00 1992
Selecting the target year
A Study on Isolated Word Recognition for Implementation of Real-Time Voice Dialing System
Lee, Hang-Seop ; Hong, Jin-U ; Lee, Gang-Seong ; Lee, Gang-Seong ;
The Journal of the Acoustical Society of Korea, volume 11, issue 1E, 1992, Pages 5~5
This paper describes speaker dependent isolated word recognition system for the development of real time voice dialing system. We used DMS model as recognition method beacuse it requires small memory for models and less computation time for matching. So we can get response in 3 seconds after utterance for 50 words vocabulary selected from department names of a university. The performance of the system showed 98 percent recognition rate for 22 sections and for 0.6 time duration weight of DMS model.
Distance Measures Based Upon Adaptive Filtering For Robust Speech Recognition In Noise
Jeong, Won Guk ; Eun, Jong Gwan ;
The Journal of the Acoustical Society of Korea, volume 11, issue 1E, 1992, Pages 15~15
On Realizing the Predictor for the Waveform Coding of Speech Signals by using the Dual First Order Autocorrelation
Lee, Mi-Suk ; Bae, Myeong-Jin ; Lee, Ju-Heon ;
The Journal of the Acoustical Society of Korea, volume 11, issue 1E, 1992, Pages 23~23
The speech waveforms are highly correlated between the adjacent samples. One way of increasing the correlation in speech signals is to simply integrate the input signals prior to coding. The integrated values can be rerooved by conventional differentiation at the receiver. This emphasizes the low frequencies of speech signals and increases the correlation between adjacent samples. The above arrano;ement is called as a sigma-delta. In this paper, we propose a new predictor which use such characteristics of sigma-delta. That is, we integrate input signals prior to coding and then, predict the present integrate sample by using two samples, one past and one next. The proposed predictor has higher mean prediction gain of 8.65dB than that of the CCITT-Recommendation ADPCM.
Active Vibration Control of a Simply Supported Plate with Piezoelectric Sensors and Actuators-II. Experiment
The Journal of the Acoustical Society of Korea, volume 11, issue 1E, 1992, Pages 30~30
Pitch Detection by Synchronizing the Phase of Noise-Corrupted Speech Signals
Lee, Byeong-Guk ; Bae, Myeong-Jin ; An, Su-Gil ;
The Journal of the Acoustical Society of Korea, volume 11, issue 1E, 1992, Pages 42~42
A new pitch detection algorithm is proposed. It takes advantage of the fact that if the phases of the fundamental and its harmonics are synchronized, the superimposed waveform shows peaks at the same peak position of the fundamental. We set the phase of the Fourier-transformed speech signal to zero, effectively synchronizing the phases of all harmonics. The algorithm gives robust performance even in 0 dB SNR environment, with a gross error rate of 3. 63%. The gross error for clean speech is only 0.18%. It also exhibits good pitch resolution since the decision logic works in the time domain. Overall experimental results indicate that the proposed algorithm is quite effective for pitch detection.
The Noise Radiation Characteristics of Axial Fan by Experimental Method
Kim, Dong-Gyu ; O, Jae-Eung ;
The Journal of the Acoustical Society of Korea, volume 11, issue 1E, 1992, Pages 50~50
Blade passage frequency tone of fan is the most obvious component among the overall noise spectrum. It is generally the most annoying component and thus needs to be reduced, Therefore, to reduce the noise level, the noise source and noise radiation characteristics identification of axial fan need to be studied in detail. In this study, noise source mechanism and noise radiation characteristics of axial fan was identified. In noise source analysis by sound pressure and sound intensity method, we carried out triggering of axial fan by photo sensor. The determination of recording time to identify the exact location of noise source on the fan blade was presented. The location of noise source exists between trailing edge of each blade and leading edge of the following blade respectively, when axial fan is rotating. We determined the noise radiation pattern of axial fan through directivity pattern and also visualized the flow of sound by vector energy flow mapping. The rotating vibration characteristics on the fan blade surface was identified by strain gauge and the coherence of structure-borne sound to sound pressure was measured as well, The possibility of static pressure measurement on the fan blade surface by piezo film was presented.
Digital Active Noise Control System Used Inverse Model
Jeong, Chan Su ; Lee, Gang Uk ; Jeong, Yang Eung ;
The Journal of the Acoustical Society of Korea, volume 11, issue 1E, 1992, Pages 56~56
The poblem of active oise control has been analysed using a adaptive signal processing technique. In this methods, the adaptive signal processor or model predicts the primary sound wave travelling along the acoustic plant and generates the secondary source
out of phase which attempts to attempts to attenuate the undesired noise by destructive interference. In the solutions presented here, acoustic propagation delay is considered as a part of the model which used the FIR filter. The effects of error path and auxiliary path transfer functioin are anayzed and a new on=-line technique for error path modeling, adaptive delayed inverse modeling is presented. In this study, using these new concepts, our system can more reduce the noise level in duct to 5dB-15dB than only using LMS algorithm system.
Auditory Evoked Responses Relating to the Interaural Crosscorrelation of Sound Field
The Journal of the Acoustical Society of Korea, volume 11, issue 1E, 1992, Pages 64~64
Speech analysis using the Robust Time-Weighted Kalman filtering
Choe, Hong-Seop ; An, Su-Gil ;
The Journal of the Acoustical Society of Korea, volume 11, issue 1E, 1992, Pages 73~73
In this paper time-varying speech signal is analyzed by using the Kalman filtering methods. In general assuming that the speech process could be stationary in short time duration, frame-based analysis method, such as LPC(Linear Predictive Coding), SSLPC(Sample Selective LPC), has been utilized to obtain the useful information of speech signal, which IS, however, not suitable for applying to the time-varying signal. Kalman filtering is generally considered to be an appropriate means for estimation of the time-varying AR(Autoregressive) speech model. Now we consider two limiting factors in using the conventional analysis method. First the familiar Kalman filter procedure has a infinite memory which degrades the ability of adaptive estimation of rapid changing parameter in the current speech. In addition to infinite memory effect, the second is that the sequential Kalman filtering method poorly estimates the parameter coefficients when periodic impulse trains are the excitation source, as in voiced speech. Therefore we propose the robust Kalman filter with time-weighted-error criterion which is applied to analyze the synthetic speech signal.
A Korean Large Vocabulary Speech Recognition System for Automatic Telephone Number Query Service
Gu, Jun-Mo ; Kim, Hyeong-Sun ; Eun, Jong-Gwan ;
The Journal of the Acoustical Society of Korea, volume 11, issue 1E, 1992, Pages 86~86
In this paper, we introduce a Korean large vocabulary speech recognition system which can recognize sentence utterances with vocabulary size of 1160 words, and be used for automatic telephone number query service. This system consists of four sub-systems. The first one is an acoustic processor recognizing words in an input sentence by Hidden Markov Mode(HMM) based speech recognition algorithm. The second one is a linguistic processor which estimates input sentence from the result of the acoustic processor and determines next words after a word by using the syntactic information. The third one is a time reduction processor reducing recognition time by limiting the number of candidate words to be computed in the acoustic processor. The time reduction processor uses linguistic information and acoustic information contained in an input sentence. The last one is a speaker adaptation processor which adapts parameters of the speech recognition system to new speakers as soon as possible. The last subsystem uses VQ adaptation and HMM parameter adaptation based on the spectral mapping. We also present out recent works to improve the performance of the large vocabulary speech recognition system. These works are focused on the enhancement of the acoustic processor and the time reduction processor for speaker-independent speech recognition. New approach for the speaker adaptation is also described.