• 로봇학회논문지
• /
• 제5권1호
• /
• pp.7-13
• /
• 2010

This paper presents the sound-based emotion estimation method and the growing HRI (human-robot interaction) system for a Mon-E robot. The method of emotion estimation uses the musical element based on the law of harmony and counterpoint. The emotion is estimated from sound using the information of musical elements which include chord, tempo, volume, harmonic and compass. In this paper, the estimated emotions display the standard 12 emotions including Eckman's 6 emotions (anger, disgust, fear, happiness, sadness, surprise) and the opposite 6 emotions (calmness, love, confidence, unhappiness, gladness, comfortableness) of those. The growing HRI system analyzes sensing information, estimated emotion and service log in an edutainment robot. So, it commands the behavior of the robot. The growing HRI system consists of the emotion client and the emotion server. The emotion client estimates the emotion from sound. This client not only transmits the estimated emotion and sensing information to the emotion server but also delivers response coming from the emotion server to the main program of the robot. The emotion server not only updates the rule table of HRI using information transmitted from the emotion client and but also transmits the response of the HRI to the emotion client. The proposed system was applied to a Mon-E robot and can supply friendly HRI service to users.

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 2008년도 추계학술대회
• /
• pp.98-101
• /
• 2008

The goal of this paper is to select and analyze the sound for robot which can help and serve the big retail market, as the part of research for design foundation development of Human Robot Interface. The sounds which are related with activity, such as thanks, reject, agony,absurdity, gladness, approval, were selected among targeted sounds with 2 step SD method. The result for using Quantitative analysis model-3 with psychological similar degree of selected sound was argued with 'the brightness of sound' and 'violent of sound'.

• 제어로봇시스템학회논문지
• /
• 제10권9호
• /
• pp.763-770
• /
• 2004

The important issue of emotion recognition from speech is a feature extracting and pattern classification. Features should involve essential information for classifying the emotions. Feature selection is needed to decompose the components of speech and analyze the relation between features and emotions. Specially, a pitch of speech components includes much information for emotion. Accordingly, this paper searches the relation of emotion to features such as the sound loudness, pitch, etc. and classifies the emotions by using the statistic of the collecting data. This paper deals with the method of recognizing emotion from the sound. The most important emotional component of sound is a tone. Also, the inference ability of a brain takes part in the emotion recognition. This paper finds empirically the emotional components from the speech and experiment on the emotion recognition. This paper also proposes the recognition method using these emotional components and the transition probability.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.123-125
• /
• 2004

Recently, we are getting to interest in a user friendly machine. The emotion is one of most important conditions to be familiar with people. The machine uses sound or image to express or recognize the emotion. This paper deals with the method of recognizing emotion from the sound. The most important emotional component of sound is a tone. Also, the inference ability of a brain takes part in the emotion recognition. This paper finds empirically the emotional components from the speech and experiment on the emotion recognition. This paper also proposes the recognition method using these emotional components and the transition probability.

• 한국소음진동공학회논문집
• /
• 제15권6호
• /
• pp.724-730
• /
• 2005

Literary works contain various human emotion and historical, cultural background. It is very significant to understand sound recognition and receptions represented in many literary works. This study aims to investigate the sound impression on ancient Korean Sijo( Korean Verse) involved various traditional korean emotion, which were expressed in different situations. Firstly we selected the appropriate Sijo to express sounds, and then classified the sound, analyzed the meaning of recognition to the sound. The number of 297 sounds were classified into 13 categories, and 20 emotional meanings. Especially, 'internal sadness' characterized the korean rooted emotion were more expressed than other meanings and this meaning were symbolized by the sound of wild geese and cuckoos.

• 디지털융복합연구
• /
• 제11권11호
• /
• pp.381-387
• /
• 2013

최근 인간의 감성에 반응하고, 감성을 유도하는 감성콘텐츠가 문화산업 분야에서 크게 주목을 받으면서 멀티미디어 콘텐츠가 유발하는 감성 추출에 초점이 모아지고 있다. 게다가 최근 멀티미디어 콘텐츠가 빠르고 방대하게 생산, 유통되는 흐름으로 볼 때 콘텐츠에서 유발하는 감성을 자동으로 추출하는 기법의 연구들이 주목받고 있다. 본 논문은 멀티미디어 콘텐츠의 소리 정보 중 특정 주파수대역의 볼륨 값을 활용하여 멀티미디어 콘텐츠 내의 감성지수를 추출하는 방법에 대해 연구하고자 한다. 이러한 연구는 동영상 콘텐츠의 감성지수를 자동으로 추출할 수 있도록 하며 추출된 정보를 활용하여 사용자의 현재 감성, 혹은 날씨 등과 같은 기타 요소에 맞추어 사용자에게 맞춤형 콘텐츠를 제공하는데 사용되어질 것이다.

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 1999년도 춘계학술발표논문집 논문집
• /
• pp.265-269
• /
• 1999

This study was carried out to evaluate human subjective sensation for fabric rustling sounds and predict the sensation with quantified sound color measurements and mechanical properties of fabrics. Thirty subjects at Virginia Tech were asked to evaluate seven sound sensation descriptors by semantic differential scale after hearing eight different fabric sounds. Sound measurements were quantified by calculating total sound pressure(LPT), level range(ΔL), and frequency difference (Δf). Mechanical properties of fabrics were measured by KES-FB. Subjective sensation for fabric sound showed significant differences among fabrics except clearness sensation. Subjective sensation predicted by sound measurements showed well fitted regression equation with ΔL and LPT> Sharpness, clearness, and highness were significantly released with mechanical properties of fabrics. All sensation was found to be predicted with sound measurements L:PT and Δf and mechanical properties such as LT, 2HG5, WC, T, and W.

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 2011년도 추계학술대회
• /
• pp.43-44
• /
• 2011

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 2012년도 춘계학술대회
• /
• pp.41-42
• /
• 2012

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2011년도 춘계학술발표대회
• /
• pp.689-691
• /
• 2011

Due to the rapid development of Information and communication, the technology of multimedia presentation technology is evolving into the service that user can actively, realistically enjoy and play based on user's preference and taste not only for User's passive service. Especially, the industry related the realistic multimedia service that supports targeting Human emotion with the property of Human hearing is expected to be formed of the high value-added premium market. Audio technology is affected on human's emotion and the viewing environment around than video technology. Also the audio technology compared to video technology is a research part that appeals to human emotion and emphasize on psychological aspects. With this viewpoint, the development of intelligent and realistic audio technology needs highly specialty. In this study, "intelligent real-sound presentation technology" that support high quality and realistic audio and the "core technologies" that are composing of this will be introduced.