User Tools

Site Tools


deepperformancediff

====== Differences ====== This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
deepperformancediff [2015/10/25 18:30]
vviro [Try it out!]
deepperformancediff [2015/10/26 17:13] (current)
jganseman
Line 5: Line 5:
 | Code | [[https://​github.com/​peachnote/​sync-visualization|Github Link]] | | Code | [[https://​github.com/​peachnote/​sync-visualization|Github Link]] |
  
-**DeepPerformanceDiff** allows you to compare renditions of the same piece (of which the score is provided by Peachnote through IMSLP) by different performers (provided by YouTube). Differences in tempo and loudness are visualised in a graph that makes it easy to spot passages ​that are played ​markedly ​different ​from others. ​+**DeepPerformanceDiff** allows you to compare renditions of the same piece (of which the score is provided by Peachnote through IMSLP) by different performers (as found on YouTube). Differences in tempo and loudness are visualised in a graph that makes it easy to spot sections ​that any performer plays markedly ​differently ​from others. ​
  
 ===== Origins ===== ===== Origins =====
  
-This project originates from an idea that is part of the [[http://​music-connection-machine.peachnote.com/​|Music Connection Machine]] brainstorm taking place at [[http://​www.peachnote.com|Peachnote]]. A minimal mockup version had already been coded. During the HAMR@ISMIR hackday, this mockup has been extended to a fully functional prototype ​include ​more audio features (notably loudness), better search functionality,​ and a more flexible layout.+This project originates from an idea that is part of the [[http://​music-connection-machine.peachnote.com/​|Music Connection Machine]] brainstorm taking place at [[http://​www.peachnote.com|Peachnote]]. A minimal mockup version had already been coded. During the HAMR@ISMIR hackday, this mockup has been extended to a fully functional prototype ​which includes ​more audio features (notably loudness), better search functionality,​ and a more flexible layout.
  
 ===== Used Technologies and Data ===== ===== Used Technologies and Data =====
Line 21: Line 21:
 ===== Relation to MIR Research and Potential Impact ===== ===== Relation to MIR Research and Potential Impact =====
  
-Study of performance practice, and especially differences between performers, is a research topic that has already garnered some interest in the past (([[http://​www.sciencedirect.com/​science/​article/​pii/​S0004370205000196|Stamatatos & Widmer 2005]])) (([[http://​www.ncbi.nlm.nih.gov/​pmc/​articles/​PMC3842509/​|Gingras et al. 2013]])) (([[http://​www.mirlab.org/​conference_papers/​International_Conference/​ISMIR%202008/​papers/​ISMIR2008_240.pdf|Sapp 2008]])). Note that many of these papers ​focus on (weighted combinations of) tempo and loudness ​differentiation ​to distinguish different performers, which are the exact features ​that DeepPerformanceDiff is able to visualize ​in a single graph.+Study of performance practice, and especially differences between performers, is a research topic that has already garnered some interest in the past (([[http://​www.sciencedirect.com/​science/​article/​pii/​S0004370205000196|Stamatatos & Widmer 2005]])) (([[http://​www.ncbi.nlm.nih.gov/​pmc/​articles/​PMC3842509/​|Gingras et al. 2013]])) (([[http://​www.mirlab.org/​conference_papers/​International_Conference/​ISMIR%202008/​papers/​ISMIR2008_240.pdf|Sapp 2008]])). Note that many of these papers ​rely on (weighted combinations of) tempo and loudness to differentiate between or distinguish different performers. DeepPerformanceDiff visualizes these features in a single graph.
  
-With this project, we are able to make an intuitive visualization ​of the data. This does not only create possibilities for data exploration,​ but is also of practical interest. We think more specifically in the direction of performance practice education, where more and more people upload recordings of themselves playing on YouTube. ​They are now able to compare themselves to others, to professional ​recordings, or to earlier versions of themselves, which enables ​them to track their own progress. ​+This intuitive visualization does not only create possibilities for data exploration,​ but is also of practical interest. We think more specifically in the direction of performance practice education, where more and more people upload recordings of themselves playing on YouTube. ​Everyone is now able to compare themselves to others, to professional ​performers, or to earlier versions of themselves, which also enables ​one to track their own progress. ​
  
 ===== Try it out! ===== ===== Try it out! =====
Line 30: Line 30:
   * Alignment definitions (json) between IMSLP scores and YouTube videos   * Alignment definitions (json) between IMSLP scores and YouTube videos
   * Alignment Quality definitions (json) between IMSLP scores and YouTube videos   * Alignment Quality definitions (json) between IMSLP scores and YouTube videos
 +
 +The Bootstrap-wrapped version is in a separate branch in the repository called '​bootstrap'​.
  
 ===== Future Work ===== ===== Future Work =====
  
-For the study of performance ​We could most definitely think towards clustering of the tempo/​loudness profiles. On such a large body of data, this could give us information on stylistic relationships between performers, maybe allowing us to detect schools of performance or regional differences in interpretation. ​+For the study of performance, we could most definitely think towards clustering of the tempo/​loudness profiles. On such a large body of data, this could give us information on stylistic relationships between performers, maybe allowing us to detect schools of performance, teacher-student relationships, ​or regional differences in interpretation. ​
  
-Music education practice would most definitely be served by a tool that visualizes the progress of students. Professional performances already appear often on YouTube, and given that we are in the age of Instagram and Vine, we foresee no lack of uploads ​ofamateur ​performances in the near future either. The comparisons between performances by people on different levels with different abilities may provide valuable pedagogical insights for tutors. ​+Music education practice would most definitely be served by a tool that visualizes the progress of students. Professional performances already appear often on YouTube, and given that we live in the age of Instagram and Vine, we foresee no lack of uploads ​of amateur ​performances in the near future either. The comparisons between performances by people on different levels with different abilities may provide valuable pedagogical insights for tutors. ​
  
deepperformancediff.txt ยท Last modified: 2015/10/26 17:13 by jganseman