• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.5
• /
• pp.1590-1609
• /
• 2021

Laughter is one of the most important nonverbal sound that human generates. It is a means for expressing his emotions. The acoustic and contextual features of this specific sound are different from those of speech and many difficulties arise during their modeling process. During this work, we propose an audio laughter generation system based on unsupervised generative models: the autoencoder (AE) and its variants. This procedure is the association of three main sub-process, (1) the analysis which consist of extracting the log magnitude spectrogram from the laughter database, (2) the generative models training, (3) the synthesis stage which incorporate the involvement of an intermediate mechanism: the vocoder. To improve the synthesis quality, we suggest two hybrid models (LSTM-VAE, GRU-VAE and CNN-VAE) that combine the representation learning capacity of variational autoencoder (VAE) with the temporal modelling ability of a long short-term memory RNN (LSTM) and the CNN ability to learn invariant features. To figure out the performance of our proposed audio laughter generation process, objective evaluation (RMSE) and a perceptual audio quality test (listening test) were conducted. According to these evaluation metrics, we can show that the GRU-VAE outperforms the other VAE models.

• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.421-430
• /
• 2018

Many voice recognition systems use methods such as MFCC, HMM to acknowledge human voice. This recognition method is designed to analyze only a targeted sound which normally appears between a human and a device one. However, the recognition capability is limited when there is a group sound formed with diversity in wider frequency range such as dog barking and indoor sounds. The frequency of overlapped sound resides in a wide range, up to 20KHz, which is higher than a voice. This paper proposes the new recognition method which provides wider frequency range by conjugating the Wideband Sound Spectrogram and the Keras Sequential Model based on DNN. The wideband sound spectrogram is adopted to analyze and verify diverse sounds from wide frequency range as it is designed to extract features and also classify as explained. The KSM is employed for the pattern recognition using extracted features from the WSS to improve sound recognition quality. The experiment verified that the proposed WSS and KSM excellently classified the targeted sound among noisy environment; overlapped sounds such as dog barking and indoor sounds. Furthermore, the paper shows a stage by stage analyzation and comparison of the factors' influences on the recognition and its characteristics according to various levels of noise.

• The Journal of Korean Academy of Prosthodontics
• /
• v.25 no.1
• /
• pp.71-94
• /
• 1987

This study was carried out to investigate the sound distortion following the alternation of the palatal plate thickness. For this study, 2 healthy male subjects (24-year-old) were selected. Born in Seoul, they both spoke Seoul dialect. First, their sounds of /na(나)/, /da(다)/, /1a(라)/, /ja(자)/, /cha(차)/, /ta(타)/, without inserting plates were recorded, and then the sounds with palatal plates of different thickness were recorded, successively. The plate was fabricated in 3 types, each palatal thickness being 1.0mm, 2.5mm, dentoalveolar portion 2.5mm, other residual portion was 1.0mm, successively. Each type plates named B, C, D-type, in succession. Series of analysis were administered through Computer(16 bit) to analyze the sound distortions. These experiments were analyzed by the LPC (without weighting, pre-weighting, post-weighting) of the consonants, vowels portion, formant frequency of the vowels and word duration of the consonants. The findings led to the following conclusions: 1. There was no correlation of the distortion rate on the 2 informants. 2. Generally, vowels were not affected by the palatal plate thickness in the formant analysis, however, more distortion was detected in the LPC analysis, especially C, D-type plates. 3. Consonants distortion was more evident in the C, D-type plate. 4. The second formant was most disturbed and reduced in the all consonants with insertion of the palatal plate, especially C, D-type plate. 5. Word duration was shortened in the plate inserted(except /ja/, /cha/), especially C, D-type. 6. It was found that dentoalveolar, hard palatal sounds were severely distorted in plate inserted, and they were mainly affected by the dentoalveolar portion thickness. 7. There was correlation between palatal thickness and consonants quality.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.42 no.4
• /
• pp.1-7
• /
• 2019

The customer of massage chair is expanding day by day from middle age to all ages. In 2018, the market size was 700 billion KRW, an increase of 30 times over 10 years. However, most related SMEs suffer from excessive competition by the market monopoly of some major companies. In this situation, in order for a related company to survive, it is necessary to steadily research and develop new products. Founded in 2009, company L produces massage chairs for health and relaxation of customers. L's products use a sound wave vibration module that is favorable for human body, unlike other products that use vibration motor type. However, frequent breakdowns of massage chair due to the vulnerability of plate (leaf) springs, which play an important role in sound wave vibration modules, made sap its competitiveness. In this paper, we propose a method to design desirable plate spring structure by sequentially experimenting with five different plate springs. The results of this study are expected to contribute to improve the quality of plate spring and the reliability of sound wave vibration module. In the future, it is necessary to find a way to use it in the development of foot massage or scalp management device as well as continuous research to find optimal plate spring structure through various analysis.

• Korean Journal of Materials Research
• /
• v.21 no.7
• /
• pp.403-409
• /
• 2011

Micro speakers are used to reproduce sound in small electric and information and communications devices, such as cellular phones, PMPs, and MP3 players. The acoustical properties and sound quality, which are changed due to the decreased size of the speaker, are often adjusted varying the type and thickness of the diaphragm. The most widely used diaphragm material is thin polymer. It was previously reported by the author of this paper that the resonance frequency of a micro speaker is changed by the type and thickness of a polymer diaphragm. In this paper, the frequency response near the resonance frequency of a micro speaker was studied as functions of the type and thickness of the polymer diaphragm. While $R_{max}$ and $R_{DC}$ were affected by the type and thickness, an analysis of the electrical impedance curve revealed that $R_o(= R_{max}/R_{DC})$ and ${\Delta}f$ were not changed. Thus, $Q_{TS}$ which was function of $R_o$, ${\Delta}f$, and the resonance frequency, is only related to the resonance frequency. The increase of the resonance frequency led to a proportional rise of $Q_{TS}$. The change of the frequency response near the resonance frequency was not dependent on the type or thickness of the polymer diaphragm, but was affected by the resonance frequency.

• The Journal of the Korea Contents Association
• /
• v.14 no.12
• /
• pp.924-934
• /
• 2014

In the last 10 years the digital music contents market has grown rapidly. However digital music contents product and services are not managed with product planning and price policy considered customer attitude and digital music contents values. This study is to define the value properties of digital music contents services based on streaming and download as genre, price, sound quality, and usage appliance, and suggest the strategic market price and service composition of digital music contents service by customer attitudes about the value properties. The research used the conjoint analysis methodology based on the hedonic price model and collected 405 questionaries by users of Korean digital music contents services to the analysis. Hence 'sound quality' in download platform, and 'appliance' in streaming platform were the elements to evaluate the customer attitude. The results present the music contents productions and companies have to provide the differentiated services and price by the value properties of user preference in the market.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2001.05a
• /
• pp.380-384
• /
• 2001

The household reftigerator's noise which is closely related with resident environment can hardly be evaluated its sound quality using the Korean Standards. Radiation characteristics of compressor noise consist of tonal noise in low frequency range and, or narrow band noise in high frequency range. In this study, measuring method for detecting the abnormal and low-level noise in high frequency band is presented, and control method for its reduction is proposed. After installing wall which is similar to living condition, we determined a major concerning frequency band of noise. It can be found the directivity of high frequency noise radiated from compressor by using experiment and analysis. According to isolation of noise transfer path, remarkable noise reduction is achieved.

• ETRI Journal
• /
• v.33 no.6
• /
• pp.945-948
• /
• 2011

Object-based audio coding can provide new music applications with interactivity. To efficiently compress a lot of target audio objects, a subband-based parametric coding scheme has been adopted for MPEG spatial audio object coding. In this letter, the time-frequency (T/F) subband analysis structure is investigated. A reconfigured T/F structure is also proposed to enhance the generating performance of sound scenes such as 'karaoke' and 'solo' play in interactive music scenarios. From the experimental results, it was confirmed that the proposed scheme remarkably improves the SNR and sound quality.

• Proceedings of the KSR Conference
• /
• 2008.11b
• /
• pp.1413-1419
• /
• 2008

As the improvements of service quality is becoming an important issue the, interior noise level of a train is an important factor of comfortable ride. To reduce the interior noise level, noise sources of the train need to be removed. However, in case with a structure of large scale and multiple noise sources, an estimation of influences of major noise sources, with indentification of its traveling paths needs to be performed. In current state, to improve the interior noise reduction, consideration of sound transmission loss of the train body prior to manufacturing is usually performed. In this study, the sound transmission loss of the train body of new model train of seoul metro's line no.2 under test opeation is measured and modeling of the train body is performed. And train interior noise is predicted using the measured values.

• Journal of IKEEE
• /
• v.23 no.3
• /
• pp.893-898
• /
• 2019

Time-frequency representations are methods to display the magnitude or energy density of a signal on the two dimensional plane of both time and frequency. They are useful in analyzing the characteristics of time-varying signals. Music is a typical time-varying signal, and it can be analyzed by time-frequency representations. Recently, it is popular to change the sound quality by attaching a safety sounder to an instrument. It is performed to improve perception subjectively by spending little cost and modifying sound quality. In time domain, it is difficult to notify the difference between music signals with and without the sounder. But, it is easy to find the difference in frequency domain or in time-frequency domain. In this paper, the music signal from a flute with sounder is analyzed both in the frequency domain and in the time-frequency domain. It is confirmed that the frequency components in the mid-frequency range of 500~2500 are reinforced.