• Language and Information
• /
• v.7 no.2
• /
• pp.81-101
• /
• 2003

This paper introduces a computerized version of the Korean Standard Pronunciation Rules that can be used in speech engineering systems such as Korean speech synthesis and recognition systems. For this purpose, we build Petri net models for each item of the Standard Pronunciation Rules, and then integrate them into the sound conversion table. The reversion of the Korean Standard Pronunciation Rules regulates the way of matching sounds into grammatically correct written characters. This paper presents not only the sound conversion table but also the character conversion table obtained by reversely converting the sound conversion table. Malting use of these tables, we have implemented a Korean character into a sound system and a Korean sound into the character conversion system, and tested them with various data sets reflecting all the items of the Standard Pronunciation Rules to verify the soundness and completeness of our tables. The test results show that the tables improve the process speed in addition to the soundness and completeness.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.4
• /
• pp.281-286
• /
• 2004

The wear process of end mill is a so complicated process that a more reliable technique is required for the monitoring and controling the tool life and its performance. This research presents a new tool wear monitoring method based on the sound signal generated on the machining. The experiment carried out continuous-side-milling for using the high-speed steel end mill under wet condition. The sound pressure was measured at 0.5m from the cutting zone by a dynamic microphone, and was analyzed at frequency domain. The tooth passing frequency appears as a harmonics form, and end mill wear is related with the first harmonic. It can be concluded from the result that the tool wear is correlate with the intensity of the measured sound at tooth passing frequency estimation of end mill wear using sound is possible through frequency analysis at tooth passing frequency under the given circumstances.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.909-912
• /
• 2017

In implementing a computerized virtual training system, the auditory element is responsible for the human cognitive ability following visual elements. Especially, the improvement of hearing ability is closely related to the performance of the training, and it contributes to improvement of the training effect. In this paper, we propose a sound system that is necessary for constructing such a virtual training system as a test system that can use a sound source using a head related transfer function (HRTF). Functional and auditory tests were performed to evaluate system performance.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.1
• /
• pp.1-6
• /
• 2011

In this study, we investigated the effect of the simplification for high frequency region in Head-Related Transfer Function(HRTF) on the sound localization. For this purpose, HRTF was measured and analyzed. The result in the HRTF frequency characteristic of the back sound source showed that the decrease revel of high frequency was smaller than that of low frequency region, which means the possibility of simplification in the high frequency region. Simplification was performed by flattening of the high frequency amplitude characteristics with the insertion of the low-pass filter, whose cutoff frequency is given by boundary frequency. Auditory experiments were performed to evaluate the simplified HRTF. The result showed that direction perception was not influenced by the simplification of the frequency characteristics of HRTF for the error of sound localization. The rate of confusion for the front and back was not affected by the simplification of the frequency characteristics over 8kHz of HRTF. Finally, we made it clear that the sound localization was not affected by the simplification of frequency characteristics of HRTF over 8kHz.

• Korean Journal of Audiology
• /
• v.24 no.2
• /
• pp.71-78
• /
• 2020

Background and Objectives: Currently limited information is available on speech stimuli processing at the subcortical level in the recipients of cochlear implant (CI). Speech processing in the brainstem level is measured using speech-auditory brainstem response (S-ABR). The purpose of the present study was to measure the S-ABR components in the sound-field presentation in CI recipients, and compare with normal hearing (NH) children. Subjects and Methods: In this descriptive-analytical study, participants were divided in two groups: patients with CIs; and NH group. The CI group consisted of 20 prelingual hearing impairment children (mean age=8.90±0.79 years), with ipsilateral CIs (right side). The control group consisted of 20 healthy NH children, with comparable age and sex distribution. The S-ABR was evoked by the 40-ms synthesized /da/ syllable stimulus that was indicated in the sound-field presentation. Results: Sound-field S-ABR measured in the CI recipients indicated statistically significant delayed latencies, than in the NH group. In addition, these results demonstrated that the frequency following response peak amplitude was significantly higher in CI recipients, than in the NH counterparts (p<0.05). Finally, the neural phase locking were significantly lower in CI recipients (p<0.05). Conclusions: The findings of sound-field S-ABR demonstrated that CI recipients have neural encoding deficits in temporal and spectral domains at the brainstem level; therefore, the sound-field S-ABR can be considered an efficient clinical procedure to assess the speech process in CI recipients.

• Journal of Audiology & Otology
• /
• v.24 no.2
• /
• pp.71-78
• /
• 2020

Background and Objectives: Currently limited information is available on speech stimuli processing at the subcortical level in the recipients of cochlear implant (CI). Speech processing in the brainstem level is measured using speech-auditory brainstem response (S-ABR). The purpose of the present study was to measure the S-ABR components in the sound-field presentation in CI recipients, and compare with normal hearing (NH) children. Subjects and Methods: In this descriptive-analytical study, participants were divided in two groups: patients with CIs; and NH group. The CI group consisted of 20 prelingual hearing impairment children (mean age=8.90±0.79 years), with ipsilateral CIs (right side). The control group consisted of 20 healthy NH children, with comparable age and sex distribution. The S-ABR was evoked by the 40-ms synthesized /da/ syllable stimulus that was indicated in the sound-field presentation. Results: Sound-field S-ABR measured in the CI recipients indicated statistically significant delayed latencies, than in the NH group. In addition, these results demonstrated that the frequency following response peak amplitude was significantly higher in CI recipients, than in the NH counterparts (p<0.05). Finally, the neural phase locking were significantly lower in CI recipients (p<0.05). Conclusions: The findings of sound-field S-ABR demonstrated that CI recipients have neural encoding deficits in temporal and spectral domains at the brainstem level; therefore, the sound-field S-ABR can be considered an efficient clinical procedure to assess the speech process in CI recipients.

• Smart Media Journal
• /
• v.11 no.1
• /
• pp.31-37
• /
• 2022

This paper proposes a study on applying signal processing and deep learning for sound recognition that detects sounds commonly heard in daily life (Screaming, Clapping, Crowd_clapping, Car_passing_by and Back_ground, etc.). In the proposed sound recognition, several techniques related to the spectrum of sound waves, augmentation of sound data, ensemble learning for various predictions, convolutional neural networks (CNN) deep learning, and two-dimensional (2-D) data are used for improving the recognition accuracy. The proposed sound recognition technology shows that it can accurately recognize various sounds through experiments.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.20 no.11
• /
• pp.1018-1024
• /
• 2010

The printer noise consists of the noise of the various components and parts such as motor, fan and solenoid. And the human's printing sound recognition shows various aspects when the printer starts to print papers because the components operate at the same time. Especially, printers are usually installed in the quiet office room. Therefore the printing noise is related to its competitiveness in the market. The importance of the printer sound qualities is increasing and it is necessary to develop the sound quality evaluation system, so it is a key point to identify the noise source of the printer and develop the sound quality index to each component. By using this evaluation system, it is possible to evaluate the sound quality of a prototype printer compared to the already existing one. In this paper, the printer sound quality evaluation system was developed by the following steps. Firstly, the signal processing method was applied to the recorded printing sound to identity and split the noise of components. Secondly, the MLR(multiple linear regression) method and the psychoacoustics were used to develop the sound quality index. Finally, the improvement of the printer sound quality is possible by using the result of the MLR and the path analysis. The output of this research will be applied to the development of a new printer.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.872-876
• /
• 2005

We propose a real-time post processing of nonlinear amplified signal to improve voice recognition in remote talk. In the previous research, we have found the nonlinear amplification has unique advantage for both the voice activity detection and the sound localization in remote talk. However, the original signal becomes distorted due to its nonlinear amplification and, as a result, the rest of sequence such as speech recognition show less satisfactorily results. To remedy this problem, we implement a linearization algorithm to recover the voice signal's linear characteristics after the localization has been done.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.12
• /
• pp.1203-1208
• /
• 2009

Green IT based u-City means that u-City having Green IT concept. If we adopt the situation awareness or not, the processing of Green IT may be reduced. For example, if we recognize a lot of speech sound on CCTV in u-City environment, it takes a lot of processing time and cost. However, if we want recognize emergency sound on CCTV, it takes a few reduced processing cost. So, for detecting emergency state dynamically through CCTV, we propose our advanced speech recognition system. For the purpose of that, we adopt HMM (Hidden Markov Model) for feature extraction. Also, we adopt Wiener filter technique for noise elimination in many information coming from on CCTV in u-City environment.