musicinformationretrievalsystems

musicinformationretrievalsystems内容摘要：

Melodyhound/Tuneserver  System Description Query by humming system Record sound, system converts into sound wave Converts query sound wave into Parson’s Code Match by melodic contour Determine distance between query and possible results via editing distance (calculate the number of modifications necessary to turn one string into the other) Return results with smallest distance MIDIZ  Developers: Maria Cl225。 udia Reis Cavalcanti, Marcelo Trannin Machado , Alessandro de Almeida Castro Cerqueira, Nelson Sampaio Araujo Jnior and Geraldo Xex233。 o  Relevant Publications:  Cavalcanti, Maria Cl225。 udia Reis et al. MIDIZ: content based indexing and retrieving MIDI files. J. Braz. Comp. Soc. [online]. 1999, vol. 6, no. 220201102]. xtamp。 pid=S010465001999000300002amp。 lng=amp。 nrm=iso ISSN 01046500. doi: MIDIZ  System description  Database that stores, indexes, searches for and recovers MIDI files based on the description of a short musical passage  Allows for nonexact queries  Musical sequence is based on intervals between notes  Uses wavelet transform and a sliding window in the melody  Window defines a note sequence of a given size (2^k) and moves through the song note by note  Each sequence in the window is converted into a vector storing the interval distances  First note in a sequence is assigned the value 1  Values of the following notes are determined by their chromatic distance in relation to the first note  Those values are added together in pairs, and the result is converted into coordinates in the final vector  Songs in database are stored in a BD Tree, determined by Discriminator Zone Expression  Completed vector of query is submitted to tree, similar results are returned Museek  Developers: Darfforst LLP  URL:  System Description: Search by title, lyrics, tune fragment, or MIDI Uses pitch, contour, and rhythm to find matches MusicSurfer  Developers: Music Technology Group of the Universitat Pompeu Fabra  Relevant Publications: Cano et al. An IndustrialStrength Contentbased Music Remendation System, Proceedings of 28th Annual International ACM SIGIR Conference。 Salvador, Brazil 2020.  URL: MusicSurfer  System Description: Automatically extracts features from songs in database based on rhythm, instrumentation, and harmony Uses spectral analysis to determine timbre Uses those features to search for similar songs NameMyTune  Developers: Strongtooth, Inc  URL:  System Description: User hums query into microphone Results are found by other users determining what the song is Orpheus  Developers: Rainer Typke  Relevant Publications: Typke, Giannopoulos, Veltkamp, Wiering, van Oostrum, Using Transportation Distances for Measuring Melodic Similarity, ISMIR 2020  URL: Orpheus  System Description: Query can be example from database, hummed or whistled melody, or a MIDI file All queries are converted into internal database format before submission Similarity between query and results based on Earth Mover’s Distance  Two distributions are represented by signatures Distance represents the amount of “work” required to change one signature to the other Work = user defined distance between two signatures Probabilistic “Name That Song”  Developers: Eric Brochu and Nando de Freitas  Publications: Brochu, E., Freitas, ., Name That Song!: A Probabilistic Approach to Querying on Music and Text. NIPS. Neural Information Processing Systems: Natural and Synthetic 2020 (。