2016:Multiple Fundamental Frequency Estimation & Tracking
- 1 Description
- 2 Data
- 3 Evaluation
- 4 Submission Format
- 5 Time and hardware limits
- 6 Submission opening date
- 7 Submission closing date
- 8 Potential Participants
That a complex music signal can be represented by the F0 contours of its constituent sources is a very useful concept for most music information retrieval systems. There have been many attempts at multiple (aka polyphonic) F0 estimation and melody extraction, a related area. The goal of multiple F0 estimation and tracking is to identify the active F0s in each time frame and to track notes and timbres continuously in a complex music signal. In this task, we would like to evaluate state-of-the-art multiple-F0 estimation and tracking algorithms. Since F0 tracking of all sources in a complex audio mixture can be very hard, we are restricting the problem to 3 cases:
- Estimate active fundamental frequencies on a frame-by-frame basis.
- Track note contours on a continuous time basis. (as in audio-to-midi). This task will also include a piano transcription sub task.
- Track timbre on a continous time basis.
Task specific mailing list
In the past we have use a specific mailing list for the discussion of this task and related tasks. This year, however, we are asking that all discussions take place on the MIREX "EvalFest" list. If you have an question or comment, simply include the task name in the subject heading.
The 2009 Multi-F0 dataset will be reused, which is composed of:
- A woodwind quintet transcription of the fifth variation from L. van Beethoven's Variations for String Quartet Op.18 No. 5. Each part (flute, oboe, clarinet, horn, or bassoon) was recorded separately while the performer listened to the other parts (recorded previously) through headphones. Later the parts were mixed to a monaural 44.1kHz/16bits file.
- Synthesized pieces using RWC MIDI and RWC samples. Includes pieces from Classical and Jazz collections. Polyphony changes from 1 to 4 sources.
- Polyphonic piano recordings generated using a disklavier playback piano.
- 6, 30-sec clips for each polyphony (2-3-4-5) for a total of 30 examples,
- 10 30-sec polyphonic piano clips.
A development dataset can be found at: Development Set for MIREX 2007 MultiF0 Estimation Tracking Task.
In order to get the development set, we request you fill out the request form here to write a short statement about the data being used only for research. After filling out the form, you will receive an email from IMIRSEL with username and password included.
Last year a newly annotated polyphonic dataset was proposed. This dataset contains a wider range of real-world music in comparison to the old dataset used from 2009. Specifically, the new dataset contains 3 clips of piano solo, 3 clips of string quartet, 2 clips of piano quintet, and 2 clips of violin sonata (violin with piano accompaniment), all of which are selected from real-world recordings. The length of each clip is between 20 and 30 seconds. The dataset is annotated by the method described in the following paper:
Li Su and Yi-Hsuan Yang, "Escaping from the Abyss of Manual Annotation: New Methodology of Building Polyphonic Datasets for Automatic Music Transcription," in Int. Symp. Computer Music Multidisciplinary Research (CMMR), June 2015.
As also mentioned in the paper, we tried our best to calibrate the errors (mostly the mismatch between onset and offset time stamps) in the preliminary annotation by human labor. Since there are still potential errors of annotation that we didn’t find, we decide to make the data and the annotation publicly available after the announcement of MIREX result this year. Specifically, we encourage every participant to help us check the annotation. The result of each competing algorithm will be updated based on the revised annotation. We hope that this can let the participants get more detailed information about the behaviors of the algorithm performing on the dataset. Moreover, in this way we can join our efforts to create a better dataset for the research on multiple-F0 estimation and tracking.
This year, We would like to discuss different evaluation methods. From last year`s result, it can be seen that on note tracking, algorithms performed poorly when evaluated using note offsets. Below is the evaluation methods we used last year:
For Task 1 (frame level evaluation), systems will report the number of active pitches every 10ms. Precision (the portion of correct retrieved pitches for all pitches retrieved for each frame) and Recall (the ratio of correct pitches to all ground truth pitches for each frame) will be reported. A Returned Pitch is assumed to be correct if it is within a half semitone (+ - 3%) of a ground-truth pitch for that frame. Only one ground-truth pitch can be associated with each Returned Pitch. Also as suggested, an error score as described in Poliner and Ellis p.g. 5 will be calculated. The frame level ground truth will be calculated by YIN and hand corrected.
For Task 2 (note tracking), again Precision (the ratio of correctly transcribed ground truth notes to the number of ground truth notes for that input clip) and Recall (ratio of correctly transcribed ground truth notes to the number of transcribed notes) will be reported. A ground truth note is assumed to be correctly transcribed if the system returns a note that is within a half semitone (+ - 3%) of that note AND the returned note`s onset is within a 100ms range( + - 50ms) of the onset of the ground truth note, and its offset is within 20% range of the ground truth note`s offset. Again, one ground truth note can only be associated with one transcribed note.
The ground truth for this task will be annotated by hand. An amplitude threshold relative to the file/instrument will be determined. Note onset is going to be set to the time where its amplitude rises higher than the threshold and the offset is going to be set to the the time where the note`s amplitude decays lower than the threshold. The ground truth is going to be set as the average F0 between the onset and the offset of the note. In the case of legato, the onset/offset is going to be set to the time where the F0 deviates more than 3% of the average F0 through out the the note up to that point. There is not going to be any vibrato larger than a half semitone in the test data.
Different statistics can also be reported if agreed by the participants.
The audio files are encoded as 44.1kHz / 16 bit WAV files.
Command line calling format
Submissions have to conform to the specified format below:
doMultiF0 "path/to/file.wav" "path/to/output/file.F0"
- path/to/file.wav: Path to the input audio file.
- path/to/output/file.F0: The output file.
Programs can use their working directory if they need to keep temporary cache files or internal debuggin info. Stdout and stderr will be logged.
For each task, the format of the output file is going to be different:
For the first task, F0-estimation on frame basis, the output will be a file where each row has a time stamp and a number of active F0s in that frame, separated by a tab for every 10ms increments.
time F01 F02 F03 time F01 F02 F03 F04 time ... ... ... ...
which might look like:
0.78 146.83 220.00 349.23 0.79 349.23 146.83 369.99 220.00 0.80 ... ... ... ...
For the second task, for each row, the file should contain the onset, offset and the F0 of each note event separated by a tab, ordered in terms of onset times:
onset offset F01 onset offset F02 ... ... ...
which might look like:
0.68 1.20 349.23 0.72 1.02 220.00 ... ... ...
All submissions should be statically linked to all libraries (the presence of dynamically linked libraries cannot be guarenteed).
All submissions should include a README file including the following the information:
- Command line calling format for all executables and an example formatted set of commands
- Number of threads/cores used or whether this should be specified on the command line
- Expected memory footprint
- Expected runtime
- Any required environments (and versions), e.g. python, java, bash, matlab.
Time and hardware limits
Due to the potentially high number of particpants in this and other audio tasks, hard limits on the runtime of submissions are specified.
A hard limit of 24 hours will be imposed on runs. Submissions that exceed this runtime may not receive a result.