• Title/Summary/Keyword: MIREX

Search Result 11, Processing Time 0.023 seconds

MIREX - 음악분석 기술의 현황과 미래

  • Lee, Seok-Pil;Sin, Sa-Im
    • Broadcasting and Media Magazine
    • /
    • v.16 no.4
    • /
    • pp.72-83
    • /
    • 2011
  • 인터넷 환경의 대중화와 디지털 음원의 기하급수적인 증가에 따라, 음악을 시그널 레벨에서 직접 분석하여 분류하거나 의미정보를 추출하는 음악분석 및 검색 기술의 상용화에 대한 요구가 늘어나고 있다. Music Information Retrieval Evaluation eXchange(MIREX)는 음악 검색(Music Information Retrieval) 시스템과 알고리즘들의 평가를 위해 매년 개최되는 평가 대회이다. MIREX는 매년 정기적으로 대회 및 회의를 개최하면서 음악 분석 기술에 대한 관심도와 최신 기술동향 등을 보여주고 있다. 따라서, 음악 기술의 MIREX의 배경, 구성 및 현황 등을 살펴보면서 전 세계 음악분석 기술의 현황을 파악해 볼 수 있다. MIREX에 대한 소개와 MIREX에서 운영하는 task들의 현황을 설명하고, 이를 통하여 음악 분석 기술의 동향과 전망을 짚어본다.

A Design of Matching Engine for a Practical Query-by-Singing/Humming System with Polyphonic Recordings

  • Lee, Seok-Pil;Yoo, Hoon;Jang, Dalwon
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • v.8 no.2
    • /
    • pp.723-736
    • /
    • 2014
  • This paper proposes a matching engine for a query-by-singing/humming (QbSH) system with polyphonic music files like MP3 files. The pitch sequences extracted from polyphonic recordings may be distorted. So we use chroma-scale representation, pre-processing, compensation, and asymmetric dynamic time warping to reduce the influence of the distortions. From the experiment with 28 hour music DB, the performance of our QbSH system based on polyphonic database is very promising in comparison with the published QbSH system based on monophonic database. It shows 0.725 in MRR(Mean Reciprocal Rank). Our matching engine can be used for the QbSH system based on MIDI DB also and that performance was verified by MIREX 2011.

Quantitative Analysis of Residual 24 Organochlorine POPs in Sundried Salts (천일염 중 유기염소계 잔류성 유기오염물질(POPs) 잔류분석)

  • Choi, Geun-Hyoung;Park, Mi-Ran;Park, Jong-Min;Hong, Su-Myeong;Kwon, Oh-Kyoung;Park, Yun-Ki;Kim, Jin-Hyo
    • The Korean Journal of Pesticide Science
    • /
    • v.15 no.4
    • /
    • pp.502-506
    • /
    • 2011
  • Most countries have the legislation and regulation for POPs control in food. In here, we studied the quantitative analysis of 24 organochlorine POPs (${\alpha}$-HCH 1, ${\beta}$-HCH 2, ${\gamma}$-HCH 3, ${\delta}$-HCH 4, trans-chlrodane 5, 2,4'-DDE 6, ${\alpha}$-endosulfan 7, cis-chlordane 8, 2,4'-DDD 9, endrin 10, ${\beta}$-endosulfan 11, 2,4'-DDT 12, endosulfan sulfate 13, HCB 14, aldrin 15, trans-nonachlor 16, 4,4'-DDE 17, dieldrin 18, 4,4'-DDD 19, cis-nonachlor 20, 4,4'-DDT 21, heptachlor 22, heptachlor epoxide 23 and mirex 24) with GC-ECD. The retention time of analytes were ranged between 19.18 min and 34.69 min, and their peak intervals were over 0.05 min at least. LOQs were ranged 0.003 ~ 0.033 ng/g, and their recovery rates were showed 60.9 ~ 120.8% on the 0.1 ng/g concentration of 24 organochlorine POPs. All tested 30 sundried salts were collected on Korean retailed market, and any analyte was not found in all the samples on LOQ levels.

Extracting Predominant Melody from Polyphonic Music using Harmonic Structure (하모닉 구조를 이용한 다성 음악의 주요 멜로디 검출)

  • Yoon, Jea-Yul;Lee, Seok-Pil;Seo, Kyeung-Hak;Park, Ho-Chong
    • Journal of the Institute of Electronics Engineers of Korea SP
    • /
    • v.47 no.5
    • /
    • pp.109-116
    • /
    • 2010
  • In this paper, we propose a method for extracting predominant melody of polyphonic music based on harmonic structure. Since polyphonic music contains multiple sound sources, the process of melody detection consists of extraction of multiple fundamental frequencies and determination of predominant melody using those fundamental frequencies. Harmonic structure is an important feature parameter of monophonic signal that has spectral peaks at the integer multiples of its fundamental frequency. We extract all fundamental frequency candidates contained in the polyphonic signal by verifying the required condition of harmonic structure. Then, we combine those harmonic peaks corresponding to each extracted fundamental frequency and assign a rank to each after calculating its harmonic average energy. We finally run pitch tracking based on the rank of extracted fundamental frequency and continuity of fundamental frequency, and determine the predominant melody. We measure the performance of proposed method using ADC 2004 DB and 100 Korean pop songs in terms of MIREX 2005 evaluation metrics, and pitch accuracy of 90.42% is obtained.

Musical Genre Classification System based on Multiple-Octave Bands (다중 옥타브 밴드 기반 음악 장르 분류 시스템)

  • Byun, Karam;Kim, Moo Young
    • Journal of the Institute of Electronics and Information Engineers
    • /
    • v.50 no.12
    • /
    • pp.238-244
    • /
    • 2013
  • For musical genre classification, various types of feature vectors are utilized. Mel-frequency cepstral coefficient (MFCC), decorrelated filter bank (DFB), and octave-based spectral contrast (OSC) are widely used as short-term features, and their long-term variations are also utilized. In this paper, OSC features are extracted not only in the single-octave band domain, but also in the multiple-octave band one to capture the correlation between octave bands. As a baseline system, we select the genre classification system that won the fourth place in the 2012 music information retrieval evaluation exchange (MIREX) contest. By applying the OSC features based on multiple-octave bands, we obtain the better classification accuracy by 0.40% and 3.15% for the GTZAN and Ballroom databases, respectively.

Characteristics of Contamination for Persistent Organic Pollutants in Soil by Land Use (토지 이용형태별 잔류성유기오염물질의 오염특성)

  • Lee, Min-Jin;Kim, Kyoung-Soo;Yoon, Jeong-Ki;Kim, Tae-Seung;Kim, Jong-Guk
    • Journal of Korean Society of Environmental Engineers
    • /
    • v.31 no.3
    • /
    • pp.208-216
    • /
    • 2009
  • This study was performed to investigate levels of POPs in soil by land use and identify congener profiles of PCBs, PCDD/Fs in soil in Korea. Heptachlor, Aldrin, Endrin, Mirex, Toxaphene were not found in all areas. The concentrations of Diedrin, Chlordane, ${\Sigma}$DDT, HCB in soil samples were in ranged from N.D. to 12.08 ${\mu}g$/kg, from N.D. to 16.08 ${\mu}g$/kg, from N.D. to 38.19 ${\mu}g$/kg and from N.D. to 1.32 ${\mu}g$/kg. In case of PCBs, concentration were in ranged from N.D. to 172.12 ${\mu}g$/kg, and PCBs contaminated area was higher than other areas. The concentrations of PCDD/Fs were in ranged from 0 to 6.68 pg I-TEQ/g. In addition, the ${\Sigma}$PCDFs concentration in the industry area soil was higher than ${\Sigma}$PCDDs.

A Threshold Adaptation based Voice Query Transcription Scheme for Music Retrieval (음악검색을 위한 가변임계치 기반의 음성 질의 변환 기법)

  • Han, Byeong-Jun;Rho, Seung-Min;Hwang, Een-Jun
    • The Transactions of The Korean Institute of Electrical Engineers
    • /
    • v.59 no.2
    • /
    • pp.445-451
    • /
    • 2010
  • This paper presents a threshold adaptation based voice query transcription scheme for music information retrieval. The proposed scheme analyzes monophonic voice signal and generates its transcription for diverse music retrieval applications. For accurate transcription, we propose several advanced features including (i) Energetic Feature eXtractor (EFX) for onset, peak, and transient area detection; (ii) Modified Windowed Average Energy (MWAE) for defining multiple small but coherent windows with local threshold values as offset detector; and finally (iii) Circular Average Magnitude Difference Function (CAMDF) for accurate acquisition of fundamental frequency (F0) of each frame. In order to evaluate the performance of our proposed scheme, we implemented a prototype music transcription system called AMT2 (Automatic Music Transcriber version 2) and carried out various experiments. In the experiment, we used QBSH corpus [1], adapted in MIREX 2006 contest data set. Experimental result shows that our proposed scheme can improve the transcription performance.

Development of Audio Melody Extraction and Matching Engine for MIREX 2011 tasks

  • Song, Chai-Jong;Jang, Dalwon;Lee, Seok-Pil;Park, Hochong
    • Proceedings of the Korean Society of Broadcast Engineers Conference
    • /
    • 2012.07a
    • /
    • pp.164-166
    • /
    • 2012
  • In this paper, we proposed a method for extracting predominant melody of polyphonic music based on harmonic structure. Harmonic structure is an important feature parameter of monophonic signal that has spectral peaks at the integer multiples of its fundamental frequency. We extract all fundamental frequency candidates contained in the polyphonic signal by verifying the required condition of harmonic structure. Then, we combine those harmonic peaks corresponding to each extracted fundamental frequency and assign a rank to each after calculating its harmonic average energy. We run pitch tracking based on the rank of extracted fundamental frequency and continuity of fundamental frequency, and determine the predominant melody. For the query by singing/humming (QbSH) task, we proposed Dynamic Time Warping (DTW) based matching engine. Our system reduces false alarm by combining the distances of multiple DTW processes. To improve the performance, we introduced the asymmetric sense, pitch level compensation, and distance intransitiveness to DTW algorithm.

  • PDF