• The Journal of the Acoustical Society of Korea
• /
• v.22 no.8
• /
• pp.711-716
• /
• 2003

This paper proposes the method to select the number of optimal mixtures in each state in Continuous Density HMM (Hidden Markov Models), Previously, researchers used the same number of mixture components in each state of HMM regardless spectral characteristic of speaker, To model each speaker as accurately as possible, we propose to use a different number of mixture components for each state, Selection of mixture components considered the probability value of mixture by each state that affects much parameter estimation of continuous density HMM, Also, we use PCA (principal component analysis) to reduce the correlation and obtain the system' stability when it is reduced the number of mixture components, We experiment it when the proposed method used average 10% small mixture components than the conventional HMM, When experiment result is only applied selection of mixture components, the proposed method could get the similar performance, When we used principal component analysis, the feature vector of the 16 order could get the performance decrease of average 0,35% and the 25 order performance improvement of average 0.65%.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.8
• /
• pp.24-30
• /
• 2001

This article is concerned with automatic segmentation for Korean speech signals. All kinds of transition cases of phonetic units are classified into 3 types and different strategies for each type are applied. The type 1 is the discrimination of silence, voiced-speech and unvoiced-speech. The histogram analysis of each indicators which consists of wavelet coefficients and SVF (Spectral Variation Function) in wavelet coefficients are used for type 1 segmentation. The type 2 is the discrimination of adjacent vowels. The vowel transition cases can be characterized by spectrogram. Given phonetic transcription and transition pattern spectrogram, the speech signal, having consecutive vowels, are automatically segmented by the template matching. The type 3 is the discrimination of vowel and voiced-consonants. The smoothed short-time RMS energy of Wavelet low pass component and SVF in cepstral coefficients are adopted for type 3 segmentation. The experiment is performed for 342 words utterance set. The speech data are gathered from 6 speakers. The result shows the validity of the method.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.1
• /
• pp.38-42
• /
• 2005

CMOS analog lowpass filters using speech signal bandwidth for a Asymmetrical Digital Subscriver Line(ADSL) modem are presented. Designed active lowpass filters are composed of the CMOS complementary high-swing cascode stage which can increase transconductance of an active element. As a result, their cutoff frequency are 138kHz and 1,100kHz respectively. A low-voltage high-swing cascode integrator which improved on a gain and unit gain frequency used to design the filters. The designed filters are verified by HSPICE simulation with the $0.251{\mu}m\;CMOS\;n-well$ Parameter and a single 2.5V power supply.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.1
• /
• pp.32-38
• /
• 2007

In order to establish the environmental condition of sounds in the classroom, we measured the impulse response in cases of using and not-using PA(Public-Address). By calculating the physical index of acoustics, I examined the differences between the two cases. The degree of improvement in listening with the help of PA has also been studied by testing the voice articulation with the use of the measured impulse response. As a result, I found that the clearness is enhanced by increasing the sound pressure level in the case of short reverberation. However, it was not the case when the reverberation time was long.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.5
• /
• pp.188-194
• /
• 2014

The feature extraction mechanism of the CASA(Computational Auditory Scene Analysis) system uses time continuity and frequency channel similarity to compose a correlogram of auditory elements. In segmentation, we compose a binary mask by using cross-correlation function, mask 1(speech) has the same periodicity and synchronization. However, when there is delay between autocorrelation signals with the same periodicity, it is determined as a speech, which is considered to be a drawback. In this paper, we proposed an algorithm to improve discrimination of channel similarity using Weighted Cross-correlation in segmentation. We conducted experiments to evaluate the speech segregation performance of the CASA system in background noise(siren, machine, white, car, crowd) environments by changing SNR 5dB and 0dB. In this paper, we compared the proposed algorithm to the conventional algorithm. The performance of the proposed algorithm has been improved as following: improvement of 2.75dB at SNR 5dB and 4.84dB at SNR 0dB for background noise environment.

• Economic and Environmental Geology
• /
• v.28 no.5
• /
• pp.487-491
• /
• 1995

The multipulse model of linear predictive coding (LPC), which has been successfully used for compressing of speech signals into an impulse excitation, is here applied to seismic data which contains multiples. Multiples are happened by successive reflection between layers and make the seismic interpretation difficult In this paper, the author applied the enhanced multipulse method to seismic traces to compress source-wavelets into spikes, and to eliminate/reduce multiples. The enhanced multipulse method which was applied to seismic traces extracted the amplitudes and locations of reflectivity function, which depicts the subsurface configuration, by iterative computation of autoregressive (AR) estimation method.

• Journal of Digital Convergence
• /
• v.12 no.6
• /
• pp.271-276
• /
• 2014

In a speech coding system using excitation source of voiced and unvoiced, it would be involved a distortion of speech quality in case coexist with a voiced and unvoiced consonants in a frame. In this paper, I propose a method of 8kbps Multi-Pulse Speech Coding(FBD-MPC: Frequency Band Division MPC) by using TSIUVC(Transition Segment Including Unvoiced Consonant) searching, extraction and approximation-synthesis method in a frequency domain. I evaluate the 8kbps MPC and FBD-MPC. As a result, SNRseg of FBD-MPC was improved 0.5dB for female voice and 0.2dB for male voice respectively. Compared to the MPC, SNRseg of FBD-MPC has been improved that I was able to control the distortion of the speech waveform finally. And so, I expect to be able to this method for cellular phone and smart phone using excitation source of low bit rate.

• Journal of the Institute of Convergence Signal Processing
• /
• v.4 no.4
• /
• pp.30-39
• /
• 2003

In this paper, we carried out the performance evaluation of HM-Net(Hidden Markov Network) speech recognition system for Korean speech databases. We adopted to construct acoustic models using the HM-Nets modified by HMMs(Hidden Markov Models), which are widely used as the statistical modeling methods. HM-Nets are carried out the state splitting for contextual and temporal domain by PDT-SSS(Phonetic Decision Tree-based Successive State Splitting) algorithm, which is modified the original SSS algorithm. Especially it adopted the phonetic decision tree to effectively express the context information not appear in training speech data on contextual domain state splitting. In case of temporal domain state splitting, to effectively represent information of each phoneme maintenance in the state splitting is carried out, and then the optimal model network of triphone types are constructed by in the parameter. Speech recognition was performed using the one-pass Viterbi beam search algorithm with phone-pair/word-pair grammar for phoneme/word recognition, respectively and using the multi-pass search algorithm with n-gram language models for sentence recognition. The tree-structured lexicon was used in order to decrease the number of nodes by sharing the same prefixes among words. In this paper, the performance evaluation of HM-Net speech recognition system is carried out for various recognition conditions. Through the experiments, we verified that it has very superior recognition performance compared with the previous introduced recognition system.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.2
• /
• pp.40-45
• /
• 2000

In speech corpus generation and speech recognition, it is sometimes needed to segment the input speech data without any prior knowledge. A method to accomplish this kind of segmentation, often called as blind segmentation, or acoustic segmentation, is to find boundaries which minimize the Euclidean distances among the feature vectors of each segments. However, the use of this metric alone is prone to errors because of the fluctuations or variations of the feature vectors within a segment. In this paper, we introduce the principal component analysis method to take the trend of feature vectors into consideration, so that the proposed distance measure be the distance between feature vectors and their projected points on the principal components. The proposed distance measure is applied in the LBDP(level building dynamic programming) algorithm for an experimentation of continuous speech segmentation. The result was rather promising, resulting in 3-6% reduction in deletion rate compared to the pure Euclidean measure.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.5
• /
• pp.441-446
• /
• 2009

This paper proposes an automatic indexing algorithm of golf video using audio information. In the proposed algorithm, the input audio stream is demultiplexed into the stream of video and audio. By means of Adaboost-cascade classifier, the continuous audio stream is classified into announcer's speech segment recorded in studio, music segment accompanied with players' names on TV screen, reaction segment of audience according to the play, reporter's speech segment with field background, filed noise segment like wind or waves. And golf swing sound including drive shot, iron shot, and putting shot is detected by the method of impulse onset detection and modulation spectrum verification. The detected swing and applause are used effectively to index action or highlight unit. Compared with video based semantic analysis, main advantage of the proposed system is its small computation requirement so that it facilitates to apply the technology to embedded consumer electronic devices for fast browsing.