• 말소리와 음성과학
• /
• 제1권1호
• /
• pp.63-67
• /
• 2009

In this paper, in order to protect music copyrights, we propose a music identification system which is scalable to the number of pieces of registered music and robust to signal-level variations of registered music. For its implementation, we define the new concepts of 'music word' and 'music phoneme' as recognition units to construct 'music acoustic models'. Then, with these concepts, we apply the HMM-based framework used in continuous speech recognition to identify the music. Each music file is transformed to a sequence of 39-dimensional vectors. This sequence of vectors is represented as ordered states with Gaussian mixtures. These ordered states are trained using Baum-Welch re-estimation method. Music files with a suspicious copyright are also transformed to a sequence of vectors. Then, the most probable music file is identified using Viterbi algorithm through the music identification network. We implemented a music identification system for 1,000 MP3 music files and tested this system with variations in terms of MP3 bit rate and music speed rate. Our proposed music identification system demonstrates robust performance to signal variations. In addition, scalability of this system is independent of the number of registered music files, since our system is based on HMM method.

• 스마트미디어저널
• /
• 제9권3호
• /
• pp.90-97
• /
• 2020

인터넷과 스마트폰의 발달로 디지털 음원은 쉽게 접근이 가능해졌고 이에 따라 음악 검색 및 추천에 대한 관심이 높아지고 있다. 음악 추천 방법으로는 장르나 감정을 분류하기 위해 음정, 템포, 박자 등의 멜로디를 사용한 연구가 진행되고 있다. 하지만 음악에서 가사는 인간의 감정을 표현하는 수단 중의 하나로 역할 비중이 점점 높아지고 있기 때문에 가사를 기반으로 한 감정 분류 연구가 필요하다. 이에 본 논문에서는 가사를 기반으로 이별 감정을 세분화하기 위해 이별 가사의 감정을 분석한다. 이별 가사에 나타나는 단어 간 유사도를 Word2Vec 학습을 통해 벡터화하여 감정 사전을 구축 한 후 LSTM을 활용하여 가사를 학습시켜 유사한 감정으로 가사를 분류해주는 Word2Vec과 LSTM을 활용한 이별 가사 감정 분류 방법을 제안한다.

• 인간행동과 음악연구
• /
• 제8권2호
• /
• pp.65-75
• /
• 2011

본 연구는 성악과 기악 배경음악이 단어 암기 과제에 미치는 영향을 알아보고자 하는데 목적이 있다. 30명의 대학생과 대학원생이 참여하여 두 그룹으로 무작위로 나뉘어 졌다. 단어 암기 과제를 하는 동안 배경음악이 삽입 되었다. 그룹 A에 있는 참가자들은 성악곡을 감상했고 그룹 B에 있는 참가자들은 동일 음악이지만 첼로로 연주되는 기악 음악을 감상 했다. 모든 참가들은 배경 음악이 있는 단어 암기 과제 후 암기력을 측정하기 위해 사후 검사가 실시되었다. T test를 통해 두 그룹의 차이를 보고자 하였다. 그룹 A의 평균(8.93)이 그룹 B(7.46) 보다 높았지만 두 그룹간의 유의미한 차이는 없었다(t(28) = 1.38, p < .05). 이는 가사가 단어 암기 과제에 방해되지 않는다는 것을 제시하였으며 배경음악이 인지 능력에 직접적인 영향을 미치지 않는다는 것을 시사한다.

• 감성과학
• /
• 제16권3호
• /
• pp.363-372
• /
• 2013

폭소노미 (foxonomy) 분위기 태그를 이용한 음악 검색 시 내부적으로 단어 태그 대신에 수치 태그 (AV 태그: Arousal과 Valence 값으로 이루어진 태그)를 이용하면 폭소노미의 문제점 중의 하나인 유사어 문제점을 일부 해결할 수 있다. 하지만 이를 위해서는 두 가지 선행 작업이 제대로 이루어져야 하는데, 그 첫 번째가 단어 태그를 수치 태그로 변환하는 작업이며 그 두 번째가 검색 대상인 음악을 수치 태그로 표현하는 작업이다. 첫 번째 작업에 대해서는 이전 연구를 통하여 그 유의성을 보였기 때문에 본 논문에서는 두 번째 작업에 대해서 그 유의성을 밝히고자 하였다. 이를 위하여 본 논문에서는 음악과 AV값 간의 관계를 정의하는 음악-분위기 매핑테이블을 제안하고, ANOVA 검증을 이용하여 분석 하였다. 실험 결과, 동의어 포함 유무에 무관하게 음악 구간의 A값과 V값 모두 12개 음악의 분위기에 대하여 분포차가 발생하고, 모두 제 1종 오류확률 P<0.001를 만족하였다. 결론적으로 음악의 분위기에 따라 AV 값 분포가 다르다는 것을 확인할 수 있었다.

• 셀메드
• /
• 제8권2호
• /
• pp.9.1-9.3
• /
• 2018

The purpose of this article is to discuss that ajaeng sanjo music playing (ASMP) can change a person's life. Ajaeng is often called a Korean cello and is the largest and lowest-pitched instrument in the string instrument family in Korean music. From far away it has a deep, low sound. Its low sound adds to its comfort and peace. Listening to a low tone music can help you calm down and heart-easing. As soon as he listened ajaeng sanjo music he felt that low sound is mildly under his patronage. He felt much more interesting energy and vitality about our minds and bodies are in ajaeng sanjo music. He could immerse himself in playing ajaeng sanjo music (ASMP) for several years. The Korean music, tune is one of expression, meaning it is inseparable from the feeling. Many studies show that low tone music not only activates brain power, but soothes minds. Music always stood by him during difficult times. In a word, ASMP leads him to change of life and music is the language of emotion to him.

• 아동학회지
• /
• 제29권2호
• /
• pp.73-87
• /
• 2008

This research studied the effects of music experiences on the language development of toddlers in the day care setting. Subjects were 44(23 experimental group; 21 control group) 29- to 42-month-old children. The experimental group experienced 10 weeks of music activities, including musical sense, singing, listening, playing instruments, improvisation, finger plays, movement, simple games, word-sound music games and music routines. Language development was tested by the Preschool Receptive-Expressive Language Scale(Kim et al. 2003). Significant differences were found in the receptive language development of 30- to 36-month-old and in the expressive language development of 36- to 42-month-old toddlers, respectively.

• 전기전자학회논문지
• /
• 제14권4호
• /
• pp.305-311
• /
• 2010

Most recent web search engines perform effective keyword-based multimedia contents retrieval by investigating keywords associated with multimedia contents on the Web and comparing them with query keywords. On the other hand, most music and compilation albums provide professional artwork as cover art that will be displayed when the music is played. If the cover art is not available, then the music player just displays some dummy or random images, but this has been a source of dissatisfaction. In this paper, in order to automatically create cover art that is matched with music contents, we propose a music album cover art creation scheme based on music contents analysis and result synthesis. We first (i) analyze music contents and their lyrics and extract representative keywords, (ii) expand the keywords using WordNet and generate various queries, (iii) retrieve related images from the Web using those queries, and finally (iv) synthesize them according to the user preference for album cover art. To show the effectiveness of our scheme, we developed a prototype system and reported some results.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.989-990
• /
• 2008

Most of all popular music is made by genre and specification of music according to age group. Generally Young people of $10{\sim}20$ ages like dance and techno, But old people over 40 age like trot. In this paper, we analyzed characteristic of music which people preferred by an age group. Without relevance with age, we could confirm the factor of music which popular in all age group by analyzing. The common factor of music all of age group liked are slow word, fast beat, repeated and simple melody, and characteristic of frequency in affluent middle tone.

• 셀메드
• /
• 제7권3호
• /
• pp.11.1-11.3
• /
• 2017

The aim of this article is to argue that a valuable tool to calm impatience is to listen to Sijo. The author seeks to argue that the best way to calm impatience is to listen to Sijo. Sijo refers to a slow and mellow music in the family of Han Ak (Korean music, 韓樂). The term slow is a revered keyword in our culture. "Slow" is a blank word in Han Ak (Korean music, 韓樂). The soul of Wol-ha Kim's Sijo is a beauty of space and easiness. Therefore, her voice will help relax the muscles of the listener and calm the soul. It is akin to the struggles of modern people competing for something excessively but ending up with nothing. We often find that gentle jogging is better than sprinting. Slow music is thus good for one's health. For example, we know that our skin can become beautiful and that real beauty can be obtained only when the body and mind are at rest and in comfort. Physical appearance depends on a healthy mind and body. The author believes that Sijo for music therapy is good for our mental health. If its effectiveness is confirmed after performing tests on animals and humans in an experimental study, we can use this type of music to treat patients with psychological illnesses.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2015년도 춘계 종합학술대회 논문집
• /
• pp.53-54
• /
• 2015

Optical music recognition (OMR) systems have been developed in recent years. However, chord symbols that play a role in a music sheet have been still disregarded. Therefore, we aimed to develop a proper approach to recognize these chord symbols. First, we divide the image of chord symbol into small segments in horizontal by a method based on vertical projection. Then, the optimal combination of these segments is found by using a lexicon-driven word scoring technique and a nearest neighbor classifier. The word that corresponds to the optimal combination is the result of recognition. The experiment gives an impressive result with accuracy 97.32%.