• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.10 no.1
• /
• pp.7-11
• /
• 2010

In this paper, we propose an improved speech/music discrimination method based on a feature combination and dimensionality reduction approach. To improve discrimination ability, we use a feature based on spectral duration analysis and employ the hierarchical dimensionality reduction (HDR) method to reduce the effect of correlated features. Through various kinds of experiments on speech and music, it is shown that the proposed method showed high discrimination results when compared with conventional methods.

• Journal of IKEEE
• /
• v.18 no.3
• /
• pp.362-367
• /
• 2014

Music score recognition applications running on a smartphone, which is one of the necessities of modern people, have already been released on the market. These applications have the several limitations, especially the recognition rate of printed music scores is low so that many errors occur when the score is played. The major factor to decrease the recognition rate comes from poor tilt-correction of the captured staff-line. In this paper, we propose a efficient method that can automatically shoot the printed music score through preview analysis, which increases the recognition rate via tilt-correction.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.11
• /
• pp.133-140
• /
• 2012

In this paper, we propose a music recommendation method considering users' tags by collaborative tagging in a social music site. Since collaborative tagging allows a user to add keywords chosen by himself to web resources, it provides users' preference about the web resources concretely. In particular, emotional tags which represent human's emotion contain users' musical preference more directly than factual tags which represent facts such as musical genre and artists. Therefore, to classify the tags into the emotional tags and the factual tags and to assign weighted values to the emotional tags, a tag ontology called UniTag is developed. After preprocessing the tags, the weighted tags are used to create user profiles, and the music recommendation algorithm is executed based on the profiles. To evaluate the proposed method, a conventional playcount-based recommendation, an unweighted tag-based recommendation, and an weighted tag-based recommendation are executed. Our experimental results show that the weighted tag-based recommendation outperforms other two approaches in terms of precision.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.1
• /
• pp.45-54
• /
• 2008

As mobile Web technology becomes more increasingly applicable. the mobile contents market. especially the music downloading for mobile phones, has recorded remarkable growth. In spite of this rapid growth, customers experience high levels of frustration in the process of searching for desired music contents. It affects to a re-purchasing rate of customers and also. music mubile content providers experience a decrease in the benefit. Therefore, in aspects of a customer relationship management (CRM), a new way to increase a benefit by providing a convenient shopping environment to mobile customers is necessary. As an solution for this situation, we propose a new music recommender system to enhance the customers' search efficiency by combining collaborative filtering with mobile web mining and ordinal scale based customer preferences. Some experiments are also performed to verify that our proposed system is more effective than the current recommender systems in the mobile Web.

• Journal of Digital Convergence
• /
• v.17 no.9
• /
• pp.375-381
• /
• 2019

Recently, the recommendation system has attracted the attention of users as customized recommendation services have been provided focusing on fashion, video and music. But these services are difficult to provide users with proper service according to many different contexts because they do not use contextual information emerging in real time. When applied contextual information expands dimensions, it also increases data sparsity and makes it impossible to recommend proper music for users. Trying to solve these problems, our study proposed a music recommendation system to recommend proper music in real time by applying association rules and using relationships and rules about the current location and time information of users. The accuracy of the recommendation system was measured according to location and time information through 5-fold cross validation. As a result, it was found that the accuracy of the recommendation system was improved as contextual information accumulated.

• International Journal of Contents
• /
• v.16 no.4
• /
• pp.68-77
• /
• 2020

VR (Virtual Reality) contents make the audience perceive virtual space as real through the virtual Z axis which creates a space that could not be created in 2D due to the space between the eyes of the audience. This visual change has led to the need for technological changes to sound and sound sources inserted into VR contents. However, studies to increase immersion in VR contents are still more focused on scientific and visual fields. This is because composing and producing VR sounds require professional views in two areas: sound-based engineering and computer-based interactive sound engineering. Sound-based engineering is difficult to reflect changes in user interaction or time and space by directing the sound effects, script sound, and background music according to the storyboard organized by the director. However, it has the advantage of producing the sound effects, script sound, and background music in one track and not having to go through the coding phase. Computer-based interactive sound engineering, on the other hand, is produced in different files, including the sound effects, script sound, and background music. It can increase immersion by reflecting user interaction or time and space, but it can also suffer from noise cancelling and sound collisions. Therefore in this study, the following methods were devised and utilized to produce sound for VR contents called "A Midsummer Night" so as to take advantage of each sound-making technology. First, the storyboard is analyzed according to the user's interaction. It is to analyze sound effects, script sound, and background music which is required according to user interaction. Second, the sounds are classified and analyzed as 'simultaneous sound' and 'individual sound'. Thirdly, work on interaction coding for sound effects, script sound, and background music that were produced from the simultaneous sound and individual time sound categories is done. Then, the contents are completed by applying the sound to the video. By going through the process, sound quality inhibitors such as noise cancelling can be removed while allowing sound production that fits to user interaction and time and space.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.12
• /
• pp.75-87
• /
• 2009

This paper, which is about the method of creating the audio fingerprint and comparing with the audio data, presents how to distinguish music using the characteristics of audio data. It is a process of applying the Difference of Gaussian (DoG: generally used for recognizing images) to the audio data, and to extract the music that changes radically, and to define the location of fingerprint. This fingerprint is made insensitive to the changes of sound, and is possible to extract the same location of original fingerprint with just a portion of music data. By reducing the data and calculation of fingerprint, this system indicates more efficiency than the pre-system which uses pre-frequency domain. Adopting this, it is possible to indicate the copyrighted music distributed in internet, or meta information of music to users.

• 한국정보교육학회:학술대회논문집
• /
• 2010.08a
• /
• pp.143-149
• /
• 2010

Although importance of music education for the mentally retarded children is realized, it is hard to improve study effects for those children with the existing music education methods due to cognitive disorders of the children. In the paper, we propose a system to improve music cognition ability of the mentally retarded children. The system is designed based on the existing MMCP theory. Our system has the following characteristics. First, the system can improve cognitive, physical, social and emotional development as well as development of music cognition ability. Second, the individualized music education is possible for the children. Third, more active and interactive education is possible.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.1
• /
• pp.69-77
• /
• 2021

This paper deals with the development of an application that converts the PDF music score entered by the user into a MIDI file of the chord the user wants. This application converts the PDF file into a PNG file for chord conversion when the user enters the PDF music score file and the chord which the user wants to change. After recognizing the melody of sheet music through image processing algorithm and recognizing the tempo of sheet music notes through deep learning, then the MIDI file of chord for existing sheet music is produced. The OpenCV algorithm and deep learning can recognize minim note, quarter note, eighth note, semi-quaver note, half rest, eighth rest, quarter rest, semi-quaver rest, successive notes and chord notes. The experiment shows that the note recognition rate of the music score was 100% and the tempo recognition rate was 90% or more.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.07a
• /
• pp.142-145
• /
• 2010

본 논문에서는 사용자의 12가지 감성 정보와 음악의 8가지 무드 카테고리를 기반으로 음악을 추천해주는 시스템을 구현하였다. 사용자의 감성과 음악의 무드를 기반으로 음악을 검색하기 위해 전공자 집단 5명과 비전공자 집단 13명, 총 18명으로부터 감성 히스토리 정보와 무드 분류 정보를 얻었다. 감성 히스토리 정보는 참여자가 자신의 감성 정보를 지정하고 어떤 음악을 들었는지를 나타내며, 무드 분류 정보는 각 곡이 어떤 무드를 갖는지를 나타낸다. 위에서 얻어진 정보를 바탕으로 사용자의 감성 정보를 기반으로 3가지 각기 다른 추천 알고리즘을 구현했다. 첫 번째 알고리즘은 사용자 감성 정보를 기반으로 얻어진 유사도 곡 리스트 중 1위곡의 무드 정보를 이용하여 음악을 추천한다. 두 번째 알고리즘은 첫 번째 알고리즘에서 1위곡부터 20위곡까지의 무드 정보를 이용하여 음악을 추천한다. 마지막 추천 알고리즘은 사용자 감성 정보를 기반으로 얻어진 유사도 곡 리스트를 등록된 사용자들이 가장 많이 들었던 순서대로 정렬하여 음악을 추천한다.