• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.3
• /
• pp.475-483
• /
• 2012

The characteristic features of speech signal do not vary significantly from frame to frame. Therefore, it is advisable to reduce the redundancy involved in the similar feature vectors. The objective of this paper is to search for the optimal condition of minimum redundancy and maximum relevancy of the speech feature vectors in speech recognition. For this purpose, we realize redundancy reduction by way of a vigilance parameter and investigate the resultant effect on the speaker-independent speech recognition of isolated words by using FVQ/HMM. Experimental results showed that the number of feature vectors might be reduced by 30% without deteriorating the speech recognition accuracy.

• Journal of the Institute of Convergence Signal Processing
• /
• v.4 no.2
• /
• pp.18-24
• /
• 2003

One of the most important subjects in the implementation of real time speech recognizer is to design both reliable VAD(Voice Activity Detection) and suitable speech feature vector. But, because it is difficult to calculate reliable VAD in the environment having surrounding noise, designed suitable speech feature vector may not be obtained. Solving this problem, in this paper, we implement not only short time power spectrum which is generally used but also two additive parameters, the comparison measure of spectrum density having robust property in noise and linear discriminant function using linear regression, then perform VAD by using the combination of each parameter having apt weight in other magnitudes of surrounding noise and confirm that proposed parameters show a robust characteristic in circumstances having surrounding noise by using DTW(Dynamic Time Waning) in recognition experiment.

• Journal of Broadcast Engineering
• /
• v.18 no.6
• /
• pp.919-926
• /
• 2013

The acoustic signal from a distance sound source in an enclosed space often produces reverberant sound that varies depending on room impulse response. The estimation of the level of reverberation or the quality of the observed signal is important because it provides valuable information on the condition of system operating environment. It is also useful for designing a dereverberation system. This paper proposes a speech quality estimation method based on the harmonicity of received signal, a unique characteristic of voiced speech. At first, we show that the harmonic signal modeling to a reverberant signal is reasonable. Then, the ratio between the harmonically modeled signal and the estimated non-harmonic signal is used as a measure of standard room acoustical parameter, which is related to speech clarity. Experimental results show that the proposed method successfully estimates speech quality when the reverberation time varies from 0.2s to 1.0s. Finally, we confirm the superiority of the proposed method in both background noise and reverberant environments.

• Speech Sciences
• /
• v.12 no.4
• /
• pp.197-202
• /
• 2005

The cochlear implantation(CI) as an useful tool for aural rehabilitation in bilateral severe to profound hearing impairment. However, CI prefer to usually one ear in spite of bilateral hearing impaired. because of the various characteristics of hearing loss, the hearing conservation for the future possibility, and socioeconomic condition of hearing impaired person and their families. The unilateral CI has limitations such as a directional loss, a difficult speech understanding in noise and a neural plasticity. These limitations will be overcome by hearing aid(HA) which is familiar with hearing impairer. but HA fitting for bimodal-binaural hearing are difficult because the difference output characteristic of HA and CI. This study will be confirm realities of use of HA in unilateral cochlear implantee. For this goal, 25(m:f=10:15) child participated who are used to HA for 1 to 17 months. We had telephone interviews with their mother about use of HA, change of auditory performance and own voice. As the results, hearing threshold levels of unimplanted ear, the use of a appropriate HA, implanted and aided hearing threshold level(HTL) are must be considered for successful biomodal-binaural hearing. Especially, implanted and aided HTL should be very useful parameter for a prediction of HA effect and a criterion of selection for bilateral cochlear implantation.

• Speech Sciences
• /
• v.15 no.4
• /
• pp.107-119
• /
• 2008

This paper intends to investigate the correlation of 12 MDVP measurements with age, sex and body-size of sampled healthy patients. In order to extract pitch and 12 MDVP parameters efficiently and display the correlation of each parameter easily, we developed the speech analysis program using C/C++ and MFC development tool. The sample group consists of 205 males and 343 females with ages 9-81. We collected vowel voices /a/ and 8 body-size measurements from them. Body-size values were taken at 8 different torso positions of each person. We analyzed the matched voice samples and body-size measurements by the developed speech analysis program and SPSS program. The result shows that a typical characteristic age-F0 pattern that F0 of male subjects are rapidly decreased after mutational period and have stable state with age and that of female subjects are slowly changed by overall age. In MDVP, age-STD in males, age-sPPQ in females relationships are especially similar to the age-F0 relationship. In case of male group, sPPQ(0.316%), Jitt(0.04%), Shim(0.25%), APQ(0.28%) variables are increased with age after mutational period. And Jitt(0.042%), sPPQ(0.219%) of females group are increased with age too. In cases of height, weight and BMI there exists a weak correlation with MDVP, which shows a correlation coefficient below 0.25 about male and female groups. The survey of correlation relationship between 8 body-size measurements and MDVP shows a insignificant statistical result by only just having the correlation coefficient maximum in M8-8 and F0(-0.394%) for males and M8-6,7(-0.368%, -0.364%) for females.

• Speech Sciences
• /
• v.11 no.3
• /
• pp.201-211
• /
• 2004

This paper introduces a development of an electrical stimulator for auditory stimulation. The electrical stimulator is useful in neurotological diagnosis, audiological evaluation, candidate selection for cochlear implantation, optimal device selection and decision making of MAP strategy for severe-to-profound hearing impaired persons. The development was based on sound parameters of auditory brainstem responses and auditory electrophysiological characteristic such as effective firing of auditory nerve and recording evoked potentials during refractory period of neuron. Besides pulse parameter could adjustable by programming for more varied electrical stimulation evoked response audiometry. Using the electrical stimulator, electrical square pulse was applied to promontory, and electrically evoked auditory brainstem response and electrically middle latency response were successfully recorded in cats.

• Phonetics and Speech Sciences
• /
• v.10 no.4
• /
• pp.147-154
• /
• 2018

Wet voice quality (WVQ) is a characteristic that appears after swallowing. Although the concept is accepted by many clinicians worldwide, it is nevertheless ambiguous. In this study, we investigated WVQ in patients with swallowing disorders using acoustic analysis. A total of 106 patients diagnosed with penetration-aspiration by the videofluoroscopic swallowing study (VFSS) were recruited. A voice recording of vowel /a/ was conducted before and after the VFSS, and an acoustic analysis was then performed using PRAAT. Voice after VFSS was used for a perceptual judgment and divided into two groups: the Wet group (48 patients) and the Non-wet group (58 patients). At the post-VFSS stage, the two groups displayed significant differences in many acoustic parameters including F0_SD, Jitter, RAP, Shimmer, APQ, HNR, NHR, FUF, DVB, and CPP. The parameter affecting judging wetness resulted into Jitter and NHR by the logistic regression test. At the pre-VFSS stage, the two groups differed significantly in many acoustic parameters including Intensity, Jitter, RAP, Shimmer, NHR, FUF, DVB, and CPP. Both pre-and post-VFSS, the mean values of all significant parameters, except Intensity, HNR, and CPP, were higher in the Wet group. According to pre-and post-VFSS, the two groups displayed interactions in many parameters (Intensity, F0_SD, Jitter, RAP, Shimmer, APQ, HNR, NHR, FUF, DVB, and CPP). In particular, Intensity increased in both groups after the VFSS, although the increase in the Non-wet group was greater. Based on these results, it was conjectured that the WVQ after swallowing resulted from the secretion effect of the mucous membrane due to the dry laryngeal characteristic of elderly patients, rather than aspiration resulting in food on the vocal cords.

• The KIPS Transactions:PartB
• /
• v.18B no.6
• /
• pp.397-404
• /
• 2011

In this paper, we were carried out experiments to apply parameter of voice analysis to measure changing characteristic articulator according to inhale the helium gas. The helium gas was used to overcome air embolism nitrogen gas to deal a fatal blow in body nitrogen gas by diver. However, the helium gas has been much trouble interpretation about abnormal voice of diver to cause squeaky voice of low articulation. Therefor, we was carried out experiments about pitch and spectrogram measurement, analysis based on to influence in acoustic organs before and after of inhaled helium gas.

• Journal of KIISE:Software and Applications
• /
• v.27 no.5
• /
• pp.583-592
• /
• 2000

This paper proposes feature parameter transformation method using independent component analysis (ICA) for speaker identification. The proposed method assumes that the cepstral vectors from various channel-conditioned speech are constructed by a linear combination of some characteristic functions with random channel noise added, and transforms them into new vectors using ICA. The resultant vector space can give emphasis to the repetitive speaker information and suppress the random channel distortions. Experimental results show that the transformation method is effective for the improvement of speaker identification system.

• Journal of Digital Convergence
• /
• v.13 no.10
• /
• pp.353-358
• /
• 2015

The Vocabulary Recognition System made by recognizing the standard vocabulary is seen as a decline of recognition when out of the standard or similar words. In this case, reconstructing the system in order to add or extend a range of vocabulary is a way to solve the problem. This paper propose configured Bhattacharyya algorithm standing by speech recognition learning model using the Bayesian methods which reflect parameter estimation upon the model configuration scalability. It is recognized corrected standard model based on a characteristic of the phoneme using the Bayesian methods for parameter estimation of the phoneme's data and Bhattacharyya algorithm for a similar model. By Bhattacharyya algorithm to configure recognition model evaluates a recognition performance. The result of applying the proposed method is showed a recognition rate of 97.3% and a learning curve of 1.2 seconds.