2008:Query by Tapping

Overview

The main purpose of QBT(Query by Tapping) is to evaluate MIR system in retrieval ground-truth MIDI files by the tapping rhythm. This task provides query rhythm files both in WAV and CSV(symbolic) formate. Evaluation database and query files can be download from http://210.68.135.13/ki/QBT.rar (revised on 2008/8/18)

Task description

Submissions to this task are expected to query a collection of wav files to retrieve 129 tapped queries. The queries are provided in both MIDI and CSV formats (time interval in MS between taps). Results should be returned in the form of a ranked list of WAV filenames for each query.

• WAV file collection: 272 WAV files.
• Queries: queries are specified in two formats:
  • Midi files: 129 mono MIDI files. So far, 1~6 human assessors have listened and tapped a 15 seconds query rhythm from beginning of each target song.
  • CSV files: 129 text files in CSV format. These contain the time interval (in ms) between first 25 NoteOn events extracted from the query Midi files, separated with comma. Rest note and time intervals smaller than 100ms have been discarded.
• Evaluation: Mean Reciprocal rank. Only the top 10 candidates for each query will be considered.

Data processing proposal for calling formats

Indexing the MIDIs collection

Indexing_exe <var1> <var2>

where

<var1>==<directory_of_MIDIs> 
<var2>==<indexing_files_output_and_working_directory>

Running for the query files

Running_exe <var3> <var4> <var5>

where

<var3>==<directory_of_indexed_file> 
<var4>==<directory_of_query_rhythm_files> 
<var5>==<answer_file_output_directory> 

Data processing for output answer file formats

The answer file for each run would look like:

Q0001:0003,0567,0999,<insert X more responses>,XXXX
Q0002:0103,0567,0998,<insert X more responses>,XXXX
Q000X:0002,0567,0999,<insert X more responses>,XXXX

Each line represents to each of the queries in a given task run.

Submission closing date

22th August 2008.

Interested Participants

• Shu-Jen Show Hsiao(show.cs95g at nctu.edu.tw)
• Rainer Typke: I would be interested if the query data can also be made available in symbolic form so we can see what part of the performance comes from good onset detection from audio, and what comes from a good matching algorithm.