Skip to content

Audio Processing Seminar 2016: Reproducible Research

Notifications You must be signed in to change notification settings

audiolabs/APSRR-2016

Repository files navigation

Reproducible Audio Research Seminar 2016

This repository presents the results of the 2016 reproducible research seminar held at International Audio Laboratories Erlangen.

The objective of this course is to understand and verify published computational results in audio research. Students presented and discussed their results of reproducing a chosen paper. Students were asked to judge on the papers reproducibility score between 1 and 6:

Score
⭐️ Cannot be reproduced.
⭐️⭐️ Cannot seem to be reproduced.
⭐️⭐️⭐️ Could be reproduced, requiring extreme effort.
⭐️⭐️⭐️⭐️ Can be reproduced, requiring considerable effort.
⭐️⭐️⭐️⭐️⭐️ Can be easily reproduced with at most 15 minutes of user effort, requiring some proprietary source packages.
🌟 Can be easily reproduced with at most 15 min of user effort, requiring only standard, freely available tools

Note: The reproducibility reports were prepared by the students. The reported Reproducibilty Score below, therefore are based on the author's own report and do not necessarily reflect the view of the International Audio Laboratories Erlangen.

Seminar Organisation:

Paper Selection

Maximum Filter Vibrato Suppression for Onset Detection

Tempo Estimation for Music Loops and a Simple Confidence Measure

Singing-Voice Separation From Monaural Recordings Using Robust Principal Component Analysis

Large-Scale Content-Based Matching of Midi and Audio Files

REpeating Pattern Extraction Technique (REPET): A Simple Method for Music/Voice Separation

Analyzing Chroma Feature Types for Automated Chord Recognition

License

About

Audio Processing Seminar 2016: Reproducible Research

Resources

Stars

Watchers

Forks

Packages

No packages published