• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06c
• /
• pp.41-44
• /
• 2011

본 연구는 현재 사용자가 청취한 음악과 청취한 순서를 기반으로 다음에 사용할 아이템을 추천하는 문제를 다룬다. 우리가 제시하는 모델은 아이템 사용 로그를 기반으로 하며, 정보검색에서 많이 사용하는 N-gram모델을 사용하여 아이템의 순서열을 추출한 후 다음에 올 확률이 높은 아이템을 학습한다. 그리고 사용자가 현재 선택한 아이템의 순서열을 기반으로 다음에 가장 들을 확률이 높은 아이템을 추천한다. 또 실 세계 데이터를 기반으로 실험하여 협업적 필터링 방식과 성능을 비교한다.

• The Journal of the Acoustical Society of Korea
• /
• v.36 no.2
• /
• pp.144-150
• /
• 2017

Musical genre is indispensible in constructing music information retrieval system, such as music search and classification. In general, the spectral characteristics of a music signal are obtained based on a subband decomposition to represent the relative distribution of the harmonic and the non-harmonic components. In this paper, we investigate the subband decomposition parameters in extracting features, which improves musical genre classification accuracy. In addition, the linear projection methods are studied to reduce the resulting feature dimension. Experiments on the widely used music datasets confirmed that the subband decomposition finer than the widely-adopted octave scale is conducive in improving genre-classification accuracy and showed that the feature-dimension reduction is effective reducing a classifier's computational complexity.

• Journal of Broadcast Engineering
• /
• v.9 no.1
• /
• pp.2-7
• /
• 2004

In this paper, we proposed the content-based music information retrieval technique using some MPEG-7 low-level descriptors. Especially, pitch information and timbral features can be applied in music genre classification, music retrieval, or QBH(Query By Humming) because these can be modeling the stochasticpattern or timbral information of music signal. In this work, we restricted the music domain as O.S.T of movie or soap opera to apply broadcasting system. That is, the user can retrievalthe information of the unknown music using only an audio clip with a few seconds extracted from video content when background music sound greeted user's ear. We proposed the audio feature set organized by MPEG-7 descriptors and distance function by vector distance or ratio computation. Thus, we observed that the feature set organized by pitch information is superior to timbral spectral feature set and IFCR(Intra-Feature Component Ratio) is better than ED(Euclidean Distance) as a vector distance function. To evaluate music recognition, k-NN is used as a classifier

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.4 s.42
• /
• pp.137-145
• /
• 2006

A user remembers a melody as not the combination of pitch and duration which is written in score but the contour which is composed of the relative pitch and duration. Because of the way of remembering a melody the previous Music Information Retrieval Systems which uses keyboard Playing or score as the main input melody are not easily acceptable in Query By Humming Systems. In this paper, we mention about the considerable checkpoints for Query By Humming System and previous researches. And we propose the feature extraction which is similar with the way of remembering a melody and similarity computation algorithms between melody in humming and melody in music. The proposed similarity computation algorithms solves the problem which can be happened when only uses the relative pitches by using relative durations.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.9 no.1
• /
• pp.67-74
• /
• 2016

In this paper, we propose a watermarking technology for a mixed music. The mixed music means recreated music that contained a number of musics in one audio clip. Royalty associated with the audio content is typically imposed by the full audio content. However, the calculation of royalties gives rise to conflict between copyright holders and users in the mixed music because it uses not full audio content but a fraction of that. To solve the conflict related with the mixed music, we propose a audio watermarking technique that inserts different watermarks for each audio in the audio that make up the mixed music. The proposed watermarking scheme might have poor SNR (signal to noise ratio) to embed to each audio clip. To overcome poor SNR problem, we used inaudible pseudo random sequence which modifies typical pseudo random sequence to canonical signed digit (CSD) form. The proposed method verifies the performance by each watermark extraction and the time internal estimation valies from the mixed music.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.10
• /
• pp.399-406
• /
• 2021

With the steady growth of the content industry, the need for research that automatically recommending content suitable for individual tastes is increasing. In order to improve the accuracy of automatic content recommendation, it is needed to fuse existing recommendation techniques using users' preference history for contents along with recommendation techniques using content metadata or features extracted from the content itself. In this work, we propose a new graph-based music recommendation method which learns an LSTM-based classification model to automatically extract appropriate tagging words from sound data and apply the extracted tagging words together with the users' preferred music lists and music metadata to graph-based music recommendation. Experimental results show that the proposed method outperforms existing recommendation methods in terms of the recommendation accuracy.

• Smart Media Journal
• /
• v.11 no.10
• /
• pp.65-75
• /
• 2022

Deep learning is used as a creative tool that could overcome the limitations of existing analysis models and generate various types of results such as text, image, and music. In this paper, we propose a method necessary to preprocess audio data using the Niko's MIDI Pack sound source file as a data set and to generate music using Bi-LSTM. Based on the generated root note, the hidden layers are composed of multi-layers to create a new note suitable for the musical composition, and an attention mechanism is applied to the output gate of the decoder to apply the weight of the factors that affect the data input from the encoder. Setting variables such as loss function and optimization method are applied as parameters for improving the LSTM model. The proposed model is a multi-channel Bi-LSTM with attention that applies notes pitch generated from separating treble clef and bass clef, length of notes, rests, length of rests, and chords to improve the efficiency and prediction of MIDI deep learning process. The results of the learning generate a sound that matches the development of music scale distinct from noise, and we are aiming to contribute to generating a harmonistic stable music.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.504-510
• /
• 2009

Recently, personalized-adaptive services became the center of interest in the world. However the services about music are not widely diffused out. That is because the analyzing of music information is more difficult than analyzing of text information. In this paper, we propose a music recommendation system which provides personalized services. The system keeps a user's listening list and analyzes it to select pieces of music similar to the user's preference. For analysis, the system extracts properties from the sound wave of music and the time when the user listens to music. Based on the properties, a piece of music is mapped into a point in the property space and the time is converted into the weight of the point. At this time, if we select and analyze the group which is selected by user frequently, we can understand user's taste. However, it is not easy to predict how many groups are formed. To solve this problem, we apply the K-means clustering algorithm to the weighted points. We modified the K-means algorithm so that the number of clusters is dynamically changed. This manner limits a diameter so that we can apply this algorithm effectively when we know the range of data. By this algorithm we can find the center of each group and recommend the similar music with the group. We also consider the time when music is released. When recommending, the system selects pieces of music which is close to and released contemporarily with the user's preference. We perform experiments with one hundred pieces of music. The result shows that our proposed algorithm is effective.

• Korean Journal of Cognitive Science
• /
• v.11 no.2
• /
• pp.11-18
• /
• 2000

The tonal music organizes the tones based on the hierarchy of the relative importance of musical tones. So listeners who are familiar with a certain musical style internalize the tonal schema of that music in mind, though they have no formal training about these organizing principles. This research examined how the western music experts who are not familiar with classical Korean music interpret classical Korean music. In the other researches about musical styles outside the western music, the superficial informations such as the distribution and the duration of tones in the experimental stimulus might be used as cues to extract the tonal schema of that music. In this study, these superficial informations were controlled in order to reduce this possibility. The results showed that the western music experts might interpret classical Korean music with the tonal schema of western music. but, thought the western music experts perceived classical Korean music on the base of the western musical framework and so their interpretation was not match with the tonal schema of classical Korean music. their interpretation was more similar to classical Korean tonal schema than the non-musicians interpretation of classical Korean music. These results suggested that the extensive experiences in a specific musical style can facilitate the listeners proper apprehension of the other musical styles.

• Journal of IKEEE
• /
• v.13 no.2
• /
• pp.159-166
• /
• 2009

So far, many researches have been done to retrieve music information using static classification descriptors such as genre and mood. Since static classification descriptors are based on diverse content-based musical features, they are effective in retrieving similar music in terms of such features. However, human emotion or mood transition triggered by music enables more effective and sophisticated query in music retrieval. So far, few works have been done to evaluate the effect of human mood transition by music. Using formal representation of such mood transitions, we can provide personalized service more effectively in the new applications such as music recommendation. In this paper, we first propose our Emotion State Transition Model (ESTM) for describing human mood transition by music and then describe a music classification and recommendation scheme based on the ESTM. In the experiment, diverse content-based features were extracted from music clips, dimensionally reduced by NMF (Non-negative Matrix Factorization, and classified by SVM (Support Vector Machine). In the performance analysis, we achieved average accuracy 67.54% and maximum accuracy 87.78%.