is automatic recognition of makam necessary for mir
play

Is automatic recognition of makam necessary for MIR? Makam - PDF document

Is automatic recognition of makam necessary for MIR? Makam information is available through meta data Such a study improves our understanding of what a makam is. Segmentation into makams/flavors is not available. It is very


  1. Is automatic recognition of makam necessary for MIR? • Makam information is available through meta data • Such a study improves our understanding of what a makam is. • Segmentation into makams/flavors is not available. It is very critical for many applications. FEATURES OF MAKAMS Aoyagi, 2001 (PhD on makam Rast, Arab music) • Intervalic structure • Pitch hierarchy • Melodic Direction Powers, 1980 1

  2. Features defined in theory • Scale, intervals and intonation of specific notes in the scale (intervalic structure) • Overall melodic progression Seyir (ascending, descending, etc.) • Typical phrases, emphasis on certain scale degrees • Hierarchy of tones and their frequency of occurrence in a piece, tonic, dominant, leading tone, etc. • Melodic range • Typical modulations, flavours • Octave relation of notes 2

  3. SCALES Uşşak-Beyati- İ sfahan Hüseyni – Muhayyer - Gülizar Rast – Pesendide – Rehavi Problems encountered in pitch histogram based processing Main problem: finding a musicologically meaningful distance measure Neva( Beyati ) Emphasis is important, Some but sometimes duration is variation misleading should be tolerated Should be discarded Width can give an idea about dynamic characteristic of a note But it can also be caused by vibrato Open to improvement ….. 3

  4. N-gram based classification Classification on audio New features that can be derived from the pitch histogram Relative frequencies of notes Neva, Hüseyni and Muhayyer 4

  5. Measuring melodic progression – Symbolic level Muhayyer Hüseyni Slope(or delta) appears to be a discriminating feature -> linked also with melodic range and emphasized degrees Formulating a low dimensional feature for overall progression Comparatively difficult on audio data 5

  6. Three main types of progressions Beyati Rast 380 380 Ascending-descending Ascending 370 370 Mid-range progression 360 360 Relative Freq. (H. Commas) Relative Freq. (H. Commas) 350 350 340 340 330 330 320 320 310 310 300 300 290 0 2 4 6 8 10 12 14 16 18 20 290 Time (percentage) 0 2 4 6 8 10 12 14 16 18 20 Time (percentage) Muhayyer 380 Descending 370 First time to 360 Relative Freq. (H. Commas) observe it on 350 actual data 340 330 320 310 300 290 0 2 4 6 8 10 12 14 16 18 20 Time (percentage) Mesauring progression on audio signals Uşşak Hüseyni Muhayyer 6

  7. When scale and overall progression is recognized, Confusion would still continue for Muhayyer and Tahir Muhayyer Tahir 380 380 370 370 360 360 Relative Freq. (H. Commas) Relative Freq. (H. Commas) 350 350 340 340 330 330 320 320 310 310 300 300 290 290 0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20 Time (percentage) Time (percentage) Emphasis of (melodies leading to) a certain note may lead to a new makam. Melodic range Example description: “The makam Rast has an ascending character and is performed mainly within the low register of the scale. The scale extends below the tonic and descents as far as Yegah (D), using the Rast tetrachord” (Aydemir, 2011) Range seems to be less discriminative and min-max is not useful either. There needs to be some weighting/filtering to get a more meaningful range information. 7

  8. Open questions • Is tetrachord/pentachord detection needed? • Is detection of dominant needed? Future goals • Segmentation into melodies and detection of emphasis notes • Segmentation into flavors • Testing all features in a makam recognition task 8

Recommend


More recommend