Giter Site home page Giter Site logo

mig-fbadata's Introduction

License: GPL v3

FBA Data Repository

This repository contains the segment annotations, assessments and other metadata of student performances for the Florida all-state auditions. The dataset is available on request from the Florida Bandmasters Association (FBA).

Usage

This repository is meant to be used with the MIG-MusicPerformanceAnalysis repository and both should be cloned at the same folder level. There are additional intermediate representations such as pitch contours which are pre-computed and stored separately. To obtain those follow the steps below:

  1. Click here to download a .zip file titled MIG-FbaData.zip.
  2. Extract the contents of the downloaded file and paste it at the same folder level as this repository.
  3. Select 'Paste and Merge' when prompted.

The downloaded 'FbaData' folder from the previous link should be merged with this cloned repository before it can be used.

Dataset Description

The data set contains student recordings for 18 different instruments covering the instrument groups woodwind, brass, and percussion. The performances are assessed by expert judges in different categories. The data is anonymized - personal identification of students and judges is impossible.

Folder Structure

The folder structure is shown below:

MIG-FbaData

--> `FBA2013` 
    --> `bandname` : folder, for a particular band, can be 'middleschool', 'concertband' or 'symphonicband'
        --> `assessments` : folder, containing the assessments for each student
            --> `student-id` : folder, for a particular student. student-id is a 5 digit number. contains annotations for the student. individual files are described in the section 'Annotations & Metadata' is this readme
        --> `musicalscores` : folder, contains instrument-wise midi / pdf / xml musical scores for the auditions
        --> `excelbandname.xlsx` : this excel sheet contains the tabulated assessments for all students for a particular band.
    ...
    ...

--> `FBA2014`
    ...
    ...
...
...

Annotations & Metadata

Each student-id folder contains the following annotations and metadata files:

*_assessments.txt

Contains the judge assessments of a student's audition. Students were assessed in different categories (ex. note accuracy) and separately for different segments of the audition (ex. technical etude). Assessments are normalized to [0,1], and -1 indicates a missing assessment. The assessments are organized into a 10 x 26 matrix where rows represent segments and columns represent categories:

Rows (10 segments):

  1. lyricalEtude
  2. technicalEtude
  3. scalesChromatic
  4. scalesMajor
  5. sightReading
  6. malletEtude
  7. snareEtude
  8. timpaniEtude
  9. readingMallet
  10. readingSnare

Columns (26 categories):

  1. articulation
  2. artistry
  3. musicalityTempoStyle
  4. noteAccuracy
  5. rhythmicAccuracy
  6. toneQuality
  7. articulationStyle
  8. musicalityPhrasingStyle
  9. noteAccuracyConsistency
  10. tempoConsistency
  11. Ab
  12. A
  13. Bb
  14. B
  15. C
  16. Db
  17. D
  18. Eb
  19. E
  20. F
  21. Gb
  22. G
  23. chromatic
  24. musicalityStyle
  25. noteAccuracyTone
  26. rhythmicAccuracyArticulation

*_instrument.txt

Contains the played instruments (ex. snare drum) and their corresponding activities (ex. etude). The data is formatted as a N*2 matrix, N is the number of segments. The first column is the instrument name (str) represented by the first three letters (ex. xylophone as xyl, snare drum as sna). The second column is the activity name (str), which shares the same pool as the 26 categories described above. Instrument ids: bcl Bb Clarinet tru Trumpet tsa Tenor Saxophone flu Flute axs Alto Saxophone frh French Horn obo Oboe sna snare drum xyl xylophone tim timpani

*_segment.txt

Contains the starting time (in sec) and duration (in sec) of every segment. The data is formatted as a N*2 matrix, N is the number of segments. The first column is the starting time, and the second column is the duration.

*_pyin_pitchtrack.txt

Contains the extracted pYin pitch contour for each student recording organized as a N*2 matrix, where N is the number of frames into which the recording is split. The 1st column is the timestamp of the frame in seconds and the 2nd column is the extracted f0 (fundamental frequency) in Hz. (Note that these files are commited to the repository and can be obtained by following the steps listed under the Usage section of this readme).

Contact


Alexander Lerch [email protected]

Georgia Tech Center for Music Technology 840 McMillan Street Atlanta, GA 30332

mig-fbadata's People

Contributors

ashispati avatar siddgururani avatar alexanderlerch avatar

Stargazers

Nathan Sprague avatar  avatar Yuan-Man avatar Karn Watcharasupat avatar Nithin Mahendran avatar Kazushi Nakazawa avatar  avatar

Watchers

James Cloos avatar Jason Freeman avatar  avatar Astrid Bin avatar  avatar  avatar Jiawen Huang avatar  avatar

mig-fbadata's Issues

Folder restructuring (NEW)

FBA'year'
--> concertband
  --> assessments
    -->student-id
            assessments.txt
            instrument.txt
            pitchcontour.txt
            annotation.txt
  --> musicalscores
    --> instrument
             midiscore
             xml/pdf         
   excelconcertband.xlsx

--> middleschool
--> symphonicband```

Make all the folder names & excel sheet names consistent across years and bands

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.