• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.62 no.12
• /
• pp.1750-1754
• /
• 2013

A variable bit rate speech coder is presented. The coder is based on the observation that a speech signal can be viewed as a combination of piecewise linear signals in a short time period. The encoder detects the sample points where the slope of the signal changes, which are called the inflection points in this paper. The coder transmits the location and value for the detected inflection sample, but only the location information for the noninflection samples. In the decoder, the noninflection samples are estimated with interpolation of the received information. Several factors affecting the performance of the coder have been tested through simulation. Simulation results show that the linear interpolation produces 1 ~ 5 dB improvement over the cubic spline interpolation. And the -law companding does not provide any benefit when it is applied before the inflection detection. With low threshold values in the inflection point detection, the coder shows better MOS and more than 16 dB improvement in SNR compared to the continuously variable slope delta modulation (CVSDM).

• Phonetics and Speech Sciences
• /
• v.10 no.2
• /
• pp.7-14
• /
• 2018

The paper investigates the acoustic characteristics of English fricatives in the TIMIT corpus, with a special focus on the role of voicing in rendering fricatives in American English. The TIMIT database includes 630 talkers and 2,342 different sentences, and comprises more than five hours of speech. Acoustic analyses are conducted in the domain of spectral and temporal properties by treating gender, voicing, and place of articulation as independent factors. The results of the acoustic analyses revealed that acoustic signals interact in a complex way to signal the gender, place, and voicing of fricatives. Classification experiments using a multiclass support vector machine (SVM) revealed that 78.7% of fricatives are correctly classified. The majority of errors stem from the misclassification of /θ/ as [f] and /ʒ/ as [z]. The average accuracy of gender classification is 78.7%. Most errors result from the classification of female speakers as male speakers. The paper contributes to the understanding of the effects of voicing and gender on fricatives in a large-scale speech corpus.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1991.06a
• /
• pp.15-20
• /
• 1991

In a text-to-speech system, sound units (e. q., phonemes, words, or phrases) can be concatenated together to produce required utterance. The quality of the resulting speech is dependent on factors including the phonological/prosodic contour, the quality of basic concatenation units, and how well the units join together. Thus although the quality of each basic sound unit is high, if occur the discontinuity in the concatenation part then the quality of synthesis speech is decrease. To solve this problem, a smoothing operation should be carried out in concatenation parts. But a major problem is that, as yet, no method of parameter smoothing is availalbe for joining the segment together.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.9
• /
• pp.5495-5501
• /
• 2014

Individuals with a left frontal brain injury show significant impairments in their speech ability. The aims of the present study were (1) to assess and compare the ability of speech acts comprehension and executive function between individuals with a left frontal brain injury and normal individuals, and (2) to investigate the correlation of speech act comprehension ability factors. The study's subjects were 18 individuals with a left frontal brain injury and 18 normal control adults of the same age, gender, and educational age. The following results were obtained. First, the group of individuals with a left frontal brain injury had lower speech act comprehension, executive function than the normal control group. Second, the speech act comprehension ability of the individuals with a left frontal brain injury showed a high correlation with the executive function.

• Journal of Digital Convergence
• /
• v.11 no.7
• /
• pp.209-214
• /
• 2013

Improve the recognition performance of speech recognition systems as a method for recognizing human listening skills were incorporated into the system. In noisy environments by separating the speech signal and noise, select the desired speech signal. but In terms of practical performance of speech recognition systems are factors. According to recognized environmental changes due to noise speech detection is not accurate and learning model does not match. In this paper, to improve the speech recognition feature extraction using gamma tone and learning model using acoustic model was proposed. The proposed method the feature extraction using auditory scene analysis for human auditory perception was reflected In the process of learning models for recognition. For performance evaluation in noisy environments, -10dB, -5dB noise in the signal was performed to remove 3.12dB, 2.04dB SNR improvement in performance was confirmed.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.8
• /
• pp.583-589
• /
• 2004

Proposed is a speaker normalization method based on vector quantizer for continuous speech recognition (CSR) system in which no acoustic information is made use of. The proposed method, which is an improvement of the previously reported speaker normalization scheme for a simple digit recognizer, builds up a canonical codebook by iteratively training the codebook while the size of codebook is increased after each iteration from a relatively small initial size. Once the codebook established, the warp factors of speakers are estimated by comparing exhaustively the warped versions of each speaker's utterance with the codebook. Two sets of phones are used to estimate the warp factors: one, a set of vowels only. and the other, a set composed of all the Phonemes. A Piecewise linear warping function which corresponds to the estimated warp factor is adopted to warp the power spectrum of the utterance. Then the warped feature vectors are extracted to be used to train and to test the speech recognizer. The effectiveness of the proposed method is investigated by a set of recognition experiments using the TIMIT corpus and HTK speech recognition tool kit. The experimental results showed comparable recognition rate improvement with the formant based warping method.

• Proceedings of the KSPS conference
• /
• 2000.07a
• /
• pp.1-6
• /
• 2000

• Phonetics and Speech Sciences
• /
• v.11 no.2
• /
• pp.57-64
• /
• 2019

In this paper, we construct a Korean text-to-speech system using the Merlin toolkit which is an open source system for speech synthesis. In the text-to-speech system, the HMM-based statistical parametric speech synthesis method is widely used, but it is known that the quality of synthesized speech is degraded due to limitations of the acoustic modeling scheme that includes context factors. In this paper, we propose an acoustic modeling architecture that uses deep neural network technique, which shows excellent performance in various fields. Fully connected deep feedforward neural network (DNN), recurrent neural network (RNN), gated recurrent unit (GRU), long short-term memory (LSTM), bidirectional LSTM (BLSTM) are included in the architecture. Experimental results have shown that the performance is improved by including sequence modeling in the architecture, and the architecture with LSTM or BLSTM shows the best performance. It has been also found that inclusion of delta and delta-delta components in the acoustic feature parameters is advantageous for performance improvement.

• Korean Journal of Oriental Medicine
• /
• v.13 no.1 s.19
• /
• pp.85-92
• /
• 2007

To evaluate the suitability and effectiveness of the questionnaire concerning personal properties on 'pattern of speech and activity' according to the Sasang constitution that were used in Iksan Wonkwang Oriental Medicine, we analyzed the data of 1,335 patients obtained through the electronic chart in the aspect of 'relative discrimination ability' to Sasang constitutions and 'response ratio' using statistical Package SPSS. In categories of 'speech pattern', No.2 (speak mildly and softly) was effectively discriminating Soeum type. No.4 (talkative) and No.7 (speak fast) were effective factors for the discrimination of Soyang type, though No.4 (talkative) was needed to be improved in response ratio. The category of 'activity pattern' has shown high response ratio but low discriminating power. However, No.2 (keep staying home but avoid going out) in this category was effectively discriminating Soeum type. The discriminating power of 'pattern of speech and activity' for the age group less than 20 years old was too low, so it is necessary to develop the questionnaire for the elementary to high school students as well as for the preschoolers.

• Speech Sciences
• /
• v.15 no.3
• /
• pp.39-51
• /
• 2008

Formant measurements are one of the most important factors to objectively test cross-linguistic differences among vowels produced by speakers of any given languages. However, many speech analysis softwares present erroneous estimates and some researchers use them without any verification procedures. The purposes of this paper are to examine formant measurements of complex waves which were synthesized from the average formant values of five Korean vowels using three default methods in Praat and to verify the measured values of the five vowels produced by 20 students using one of the methods. Variances along the time axis are discussed after determining absolute difference sum from the 1/3 vowel duration point. Results show that there were smaller measurement errors by the burg method. Also, greater errors were observed in the sl or lpc methods mostly caused by the inappropriate formant settings. Formant measurement deviations were greater in those vowels produced by the female students than those of the male students, which were mostly attributed to the settings for the vowels /o, u/. Formant settings can best be corrected by changing the number of formants to the number of visible dark bands on the spectrogram. Those results suggest that researchers should check the validity of the estimates from the speech analysis software. Further studies are recommended on the perception test of the original sound with the synthesized sound by the estimated formant values.