• International journal of advanced smart convergence
• /
• v.9 no.2
• /
• pp.20-27
• /
• 2020

Sound event detection in real-world environments suffers from the interference of non-stationary and time-varying noise. This paper presents an adaptive noise reduction method for sound event detection based on non-negative matrix factorization (NMF). In this paper, we proposed a deep learning model that integrates Convolution Neural Network (CNN) with Non-Negative Matrix Factorization (NMF). To improve the separation quality of the NMF, it includes noise update technique that learns and adapts the characteristics of the current noise in real time. The noise update technique analyzes the sparsity and activity of the noise bias at the present time and decides the update training based on the noise candidate group obtained every frame in the previous noise reduction stage. Noise bias ranks selected as candidates for update training are updated in real time with discrimination NMF training. This NMF was applied to CNN and Hidden Markov Model(HMM) to achieve improvement for performance of sound event detection. Since CNN has a more obvious performance improvement effect, it can be widely used in sound source based CNN algorithm.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.7
• /
• pp.1391-1396
• /
• 2012

It is introduced an approach to model for numerical analysis of a sound absorbing material that has different absorbing coefficient according to frequency. For modeling of a sound absorbing material, we tried to model by a traditional modeling method. But it had large differences on frequency domain, especially a capacitance component due to increasing of frequency. We approach to model a sound absorbing material by the Debye polarization technique with non-linear least square method. At first, we estimated parameters form a polyurethane with thickness 25 mm, then we could model a polyurethane with thickness 50 mm using same parameters. Therefor, we could find that the Debye polarization is an useful way to model sound absorbing materials.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.2 no.2
• /
• pp.241-248
• /
• 1998

The accurate measurement of the room sound level is required in environment noise control. However, it has been found that the measurement system noise always corrupts the actual noise from the sound source. In this study, a new sound level measurement technique in which the system noise is eliminated from the measured signal by the cross spectrum method, is proposed. The received signals of two measuring microphones are recorded to DAT through the pre-amplifier and digitized by A/D converter. The cross spectrum calculated from the digitized signals gives the accurate sound level since the system noise is uncorrelated with the sound source noise which we want to measure. The performance of the proposed technique is verified experimentally to be effective and the technique is found to be economic since the low cost general purpose microphone could be used in this technique.

• International journal of advanced smart convergence
• /
• v.2 no.2
• /
• pp.16-18
• /
• 2013

In this paper, we develop a method to design a flexible thin film audio systems utilizing Polyvinylidene fluoride. The system we designed showed the properties of increased transparency and sound pressure levels. As an input terminal transparent oxide thin film is adopted. In order to provide dielectric insulation, a transparent insulating oxide thin film is coated to obtain double -layered structure. In the range of visible light, the output from the output of the system showed an increased and improved sound pressure level. The piezoelectric polymer film of polyvinylidene fluoride (PVDF) is used to produce mechanical vibration due to the applied electrical voltage signal. An analog electric voltage signal is transformed into sound waves in the audio system.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1820-1825
• /
• 2004

In this paper, we propose a humanoid active audition system which detects the direction of sound and performs speech recognition using just three microphones. Compared with previous researches, this system comprises simpler algorithm and better amplifier system having advantages to increase a detectible distance of sound signal in spite of simple circuit. In order to verify our system's performance, we install the proposed active audition system to the home service robot, called Hombot II, which has been developed at the KIST (Korea Institute of Science and Technology), thus we confirm excellent performance by experimental results

• Speech Sciences
• /
• v.11 no.3
• /
• pp.175-180
• /
• 2004

In case of the hearing impairment with tinnitus and increased sound sensitivity, it is known that the patients tend to appeal the psychologically oriented social handicap rather than communication disability. The audiologist who is responsible for such patients in aural rehabilitation should pay special attention to the counseling techniques including tinnitus retain therapy (TRT), ear protector, noise generator, or specific acoustic training based on close cooperation and rapport. And then the audiologist should try to lessen their reaction to the tinnitus by using a hearing aid. This therapies tries to focus not a. total approach but a treatment to lessen the severity of tinnitus. This paper as a case report that a unilateral sharply slopped sensorineural hearing impaired person with tinnitus and increased sound sensitivity by using four channel digital signal processing (DSP) hearing aid with programming increment at low level (PILL).

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1998.03a
• /
• pp.310-315
• /
• 1998

This study includes noise automatic inspection system for washing machine auto-transmission one of modern home necessary. We effort to find and certificate sound noisesource by sound power and sound intensity, and apply to frequency analysis in vibration related sound noise. Still more we have been studying to data acquisition and programming for MS VisualBasic version 5.0. System component is below. 1) Pentium PC for data acquisition. 2) DSO for noise acquisition. 3) S/W for comparison and decision. 4) I/F Board for data communication. Wave form data through the DSO are converting to ASCII code data. The ASCII code through binary converting S/W are to be decision fitness or the badness comparison S/W. Finally, we will making noise monitoring system and automatic inspection system.

• The Journal of the Acoustical Society of Korea
• /
• v.31 no.2
• /
• pp.73-78
• /
• 2012

The photophone, Alexander Graham Bell's device for transmitting sound through light was patented in 1880. It included the transmitter modulating and reflecting strong light like sunlight to a distant receiver which produced sound. In this working of the photophone, the discovery of the sound-emitting effect under illumination was very essential. Longing for being famous in the scientific community, Bell focused on presenting various methods for producing sounds and for maximizing the loudness by performing intensive research on the photoacoustic effect. Bell's scientific research on photoacoustics was successful in establishing himself as a scientist and laid a foundation of photoacoustic analysis. And his invention became a basis for other researchers' subsequent technologies like fiber-optic communication.

• English Language & Literature Teaching
• /
• v.13 no.2
• /
• pp.129-153
• /
• 2007

When English language teachers are deciding on their priorities for teaching pronunciation, it is imperative to know what kind of differences and errors are most likely to interfere with communication, and what special problems particular first-language speakers will have with English pronunciation. In other words, phoneme discrimination skill is an integral part of speech processing for the EFL learners' learning to converse in English. Training using sound visualization technique can be effective in improving second language learners' perceptions and productions of segmental and suprasegmental speech contrasts. This study assessed the efficacy of a pronunciation training that provided visual feedback for EFL learners acquiring pitch and durational contrasts to produce and perceive English phonemic distinctions. The subjects' ability to produce and to perceive novel English words was tested in two contexts before and after training; words in isolation and words in sentences. In comparison with an untrained control group, trainees showed improved perceptual and productive performance, transferred their knowledge to new contexts, and maintained their improvement three months after training. These findings support the feasibility of learner-centered programs using sound visualization technique for English language pronunciation instruction.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.2
• /
• pp.393-399
• /
• 2001

In order to obtain the fundamental data on the underwater sound of fishes for marine ranching. The underwater sound of Red Seabream(Pagrus major) and Schlegel's black rockfish(Sebastes schfegeli) were measured, which character were summarized by FET analysis.