Piano Evaluation for Level Normalisation » History » Version 40
Version 39 (Chris Cannam, 2014-07-23 02:24 PM) → Version 40/47 (Chris Cannam, 2014-07-23 02:40 PM)
h1. Piano Evaluation for Level Normalisation
Lack of normalisation for Vamp plugin inputs is a problem when analysing quiet recordings (see #1028).
Testing using a small set of piano recordings, quickly evaluating performance across the first 30 seconds under a number of different normalisation / level management regimes.
h3. Input files
|Filename|Signal max approx|
|@31.wav@|0.57|
|@MAPS_MUS-bach_846_AkPnBcht.wav@|0.12|
|@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|0.33|
|@MAPS_MUS-scn15_7_SptkBGAm.wav@|0.13|
|@mz_333_1MINp_align.wav@|0.10|
The plugin has one internal threshold parameter, which can be lowered to find quieter notes (at the expense of course of more false positives). We don't really want to expose this (or any continuous controls) as a parameter. But we need to have approximately predictable input levels, for this threshold to be meaningful.
h3. Methods
|Name|Hg revision|Description|
|@norm@|commit:d721a17f3e14|Normalise to 0.50 max before running plugin (can't do this in plugin: it's here as the reference case)|
|@as-is@|commit:d721a17f3e14|No normalisation|
|@to-date@|commit:d9b688700819|Track max signal level _so far_, adjust each sample so that max is at 0.50|
|@r2@,@r3@,@r4@,@r5@,@r6@|commit:b5a8836dd2a4|Preprocess with "Flatten Dynamics":/projects/flattendynamics at 0.02, 0.03, 0.04, 0.05, 0.06 target RMS levels respectively|
|@s8@|commit:4ac067799e0b|With "Flatten Dynamics second attempt":/projects/flattendynamics/wiki/Wiki with max RMS targeted to 0.08|
|@t4@|commit:d67fae2bb29e|With Flatten Dynamics attempt 2a with max RMS targeted to 0.04|
|@u4@|commit:70773820e719|With Flatten Dynamics attempt 2b with max RMS targeted to 0.04|
h3. Results
Reporting only the note onset F-measure for the first 30 seconds of each piece.
|Filename|@norm@|@as-is@|@to-date@|@r2@|@r3@|@r4@|@r5@|@r6@|@s8@|@t4@|@u4@|
|@31.wav@|50|33|40|45|47|48|45|43|42|49|45|
|@MAPS_MUS-bach_846_AkPnBcht.wav@|87|15|62|64|85|87|87|86|81|86|87|
|@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|33|31|31|11|25|31|32|31|32|34|35|
|@MAPS_MUS-scn15_7_SptkBGAm.wav@|73|16|61|50|57|67|74|75|70|69|68|
|@mz_333_1MINp_align.wav@|66|3|58|42|60|64|66|63|66|63|65|
The precision (_proportion of correct onsets among detected onsets, or 1 minus the false-positive rate_) and recall (_proportion of correctly-detected onsets among all ground-truth onsets, or true-positive rate_) vary as you would hope:
* when the resulting audio level is quieter than the @norm@ case, precision is high and recall is low but the F-measure is worse than the @norm@ case
* when the resulting audio level is louder than the @norm@ case, precision is low and recall is high and the F-measure is still worse than the @norm@ case
This suggests that our threshold (which happens to be 6) is moderately well-suited to the @norm@ case, at least to optimise F-measure (this might not be the most perceptually useful measure though).
The best results (apart from @norm@) above seem to be @r5@ and @u4@. Let's try to refine the parameters for each of those and see if any patterns emerge.
h4. Flatten Dynamics fine-tuning
The adjustable parameters within @r5@, with their defaults, are
|Parameter|Description|Default|
|@historySeconds@|Length of RMS window|4.0 sec|
|@catchUpSeconds@|Length of gain slide window|0.5 sec|
|@targetRMS@|Target RMS value|0.05|
|@maxGain@|Hard limit on gain|20.0|
The @targetRMS@ is the one we have been varying across @r2@, @r3@ etc -- for @r5@ it is fixed at 0.05. We don't need to test @maxGain@ variation.
Here @r5hNcM@ represents the @r5@ method with @historySeconds@ = N and @catchUpSeconds@ = M/10. So @r5@ is the same as @r5h4c05@. The @r5@ test was run again, hence variation from above results.
|Filename|@norm@|@r5@|@r5h2c05@|@r5h5c05@|@r5h6c05@|r5h8c05|@r5h4c01@|r5h4c10@| |Filename|@norm@|@r5@|@r5h2c05@|@r5h6c05@|@r5h4c01@|r5h4c10@|
|@31.wav@|50|47|38|47|48|46| |@31.wav@|50|47|
|@MAPS_MUS-bach_846_AkPnBcht.wav@|87|87|87|87|87|88| |@MAPS_MUS-bach_846_AkPnBcht.wav@|87|87|
|@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|33|32|33|32|29|31| |@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|33|32|
|@MAPS_MUS-scn15_7_SptkBGAm.wav@|73|73|66|72|76|73| |@MAPS_MUS-scn15_7_SptkBGAm.wav@|73|73|
|@mz_333_1MINp_align.wav@|66|66|64|64|66|63|
|@mz_333_1MINp_align.wav@|66|66|
h4. For different piano template sets
The above results are all generated using four piano templates, numbered 1-3 plus @pianorwc@.
Here are results using the @norm@ and @as-is@ methods, but with different sets of piano templates: first with three templates (1-3) and then with each template in turn as the only one.
The template turns out not to make an enormous difference -- perhaps because these recordings contain nothing but piano?
|Filename|@norm@/all|@as-is@/all|@norm@/3of4|@as-is@/3of4|@norm@/1|@as-is@/1|@norm@/2|@as-is@/2|@norm@/3|@as-is@/3|@norm@/rwc|@as-is@/rwc|
|@31.wav@|50|33|51|30|50|34|44|42|50|32|56|36|
|@MAPS_MUS-bach_846_AkPnBcht.wav@|87|15|86|16|86|24|75|20|73|10|71|18|
|@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|33|31|32|32|31|22|29|31|35|34|32|28|
|@MAPS_MUS-scn15_7_SptkBGAm.wav@|73|16|71|19|71|12|68|14|72|17|70|15|
|@mz_333_1MINp_align.wav@|66|3|68|1|63|4|67|2|67|1|63|3|
h4. For "generic" template set
The above results all use template sets with only piano templates in them.
Here are results using the @norm@ and @as-is@ methods, but with the full set of instrument templates (four pianos plus all the rest).
|Filename|@norm@|@as-is@|
|@31.wav@|49|37|
|@MAPS_MUS-bach_846_AkPnBcht.wav@|79|34|
|@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|31|28|
|@MAPS_MUS-scn15_7_SptkBGAm.wav@|67|16|
|@mz_333_1MINp_align.wav@|63|5|
h4. Cross-checking with non-piano test data
The results need to be roughly comparable with those obtained from pre-normalised data using other datasets as well as the piano one. Here is a subset of the TRIOS dataset. The @norm@ result is that obtained from the plugin prior to doing this work, using pre-normalised data.
The @mirex@ result is that from the MIREX 2012 submission in MATLAB, but note that this always uses all instrument templates while the plugin results are based on selecting the "right" instrument for the piece (which is assumed to be the best, though we aren't actually testing that here).
|File|@mirex@|@norm@|@u4@|
|mozart/piano|60|64|56|
|mozart/viola|33|37|35|
|mozart/mix|51|58|55|
|mozart/clarinet|74|80|86|
|lussier/piano|45|52|63|
|lussier/mix|36|43|40|
|lussier/bassoon|43|75|80|
|lussier/trumpet|43|46|51|
|take_five/piano|61|46|69|
|take_five/mix|62|73|69|
|take_five/saxophone|78|80|84|
Lack of normalisation for Vamp plugin inputs is a problem when analysing quiet recordings (see #1028).
Testing using a small set of piano recordings, quickly evaluating performance across the first 30 seconds under a number of different normalisation / level management regimes.
h3. Input files
|Filename|Signal max approx|
|@31.wav@|0.57|
|@MAPS_MUS-bach_846_AkPnBcht.wav@|0.12|
|@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|0.33|
|@MAPS_MUS-scn15_7_SptkBGAm.wav@|0.13|
|@mz_333_1MINp_align.wav@|0.10|
The plugin has one internal threshold parameter, which can be lowered to find quieter notes (at the expense of course of more false positives). We don't really want to expose this (or any continuous controls) as a parameter. But we need to have approximately predictable input levels, for this threshold to be meaningful.
h3. Methods
|Name|Hg revision|Description|
|@norm@|commit:d721a17f3e14|Normalise to 0.50 max before running plugin (can't do this in plugin: it's here as the reference case)|
|@as-is@|commit:d721a17f3e14|No normalisation|
|@to-date@|commit:d9b688700819|Track max signal level _so far_, adjust each sample so that max is at 0.50|
|@r2@,@r3@,@r4@,@r5@,@r6@|commit:b5a8836dd2a4|Preprocess with "Flatten Dynamics":/projects/flattendynamics at 0.02, 0.03, 0.04, 0.05, 0.06 target RMS levels respectively|
|@s8@|commit:4ac067799e0b|With "Flatten Dynamics second attempt":/projects/flattendynamics/wiki/Wiki with max RMS targeted to 0.08|
|@t4@|commit:d67fae2bb29e|With Flatten Dynamics attempt 2a with max RMS targeted to 0.04|
|@u4@|commit:70773820e719|With Flatten Dynamics attempt 2b with max RMS targeted to 0.04|
h3. Results
Reporting only the note onset F-measure for the first 30 seconds of each piece.
|Filename|@norm@|@as-is@|@to-date@|@r2@|@r3@|@r4@|@r5@|@r6@|@s8@|@t4@|@u4@|
|@31.wav@|50|33|40|45|47|48|45|43|42|49|45|
|@MAPS_MUS-bach_846_AkPnBcht.wav@|87|15|62|64|85|87|87|86|81|86|87|
|@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|33|31|31|11|25|31|32|31|32|34|35|
|@MAPS_MUS-scn15_7_SptkBGAm.wav@|73|16|61|50|57|67|74|75|70|69|68|
|@mz_333_1MINp_align.wav@|66|3|58|42|60|64|66|63|66|63|65|
The precision (_proportion of correct onsets among detected onsets, or 1 minus the false-positive rate_) and recall (_proportion of correctly-detected onsets among all ground-truth onsets, or true-positive rate_) vary as you would hope:
* when the resulting audio level is quieter than the @norm@ case, precision is high and recall is low but the F-measure is worse than the @norm@ case
* when the resulting audio level is louder than the @norm@ case, precision is low and recall is high and the F-measure is still worse than the @norm@ case
This suggests that our threshold (which happens to be 6) is moderately well-suited to the @norm@ case, at least to optimise F-measure (this might not be the most perceptually useful measure though).
The best results (apart from @norm@) above seem to be @r5@ and @u4@. Let's try to refine the parameters for each of those and see if any patterns emerge.
h4. Flatten Dynamics fine-tuning
The adjustable parameters within @r5@, with their defaults, are
|Parameter|Description|Default|
|@historySeconds@|Length of RMS window|4.0 sec|
|@catchUpSeconds@|Length of gain slide window|0.5 sec|
|@targetRMS@|Target RMS value|0.05|
|@maxGain@|Hard limit on gain|20.0|
The @targetRMS@ is the one we have been varying across @r2@, @r3@ etc -- for @r5@ it is fixed at 0.05. We don't need to test @maxGain@ variation.
Here @r5hNcM@ represents the @r5@ method with @historySeconds@ = N and @catchUpSeconds@ = M/10. So @r5@ is the same as @r5h4c05@. The @r5@ test was run again, hence variation from above results.
|Filename|@norm@|@r5@|@r5h2c05@|@r5h5c05@|@r5h6c05@|r5h8c05|@r5h4c01@|r5h4c10@| |Filename|@norm@|@r5@|@r5h2c05@|@r5h6c05@|@r5h4c01@|r5h4c10@|
|@31.wav@|50|47|38|47|48|46| |@31.wav@|50|47|
|@MAPS_MUS-bach_846_AkPnBcht.wav@|87|87|87|87|87|88| |@MAPS_MUS-bach_846_AkPnBcht.wav@|87|87|
|@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|33|32|33|32|29|31| |@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|33|32|
|@MAPS_MUS-scn15_7_SptkBGAm.wav@|73|73|66|72|76|73| |@MAPS_MUS-scn15_7_SptkBGAm.wav@|73|73|
|@mz_333_1MINp_align.wav@|66|66|64|64|66|63|
|@mz_333_1MINp_align.wav@|66|66|
h4. For different piano template sets
The above results are all generated using four piano templates, numbered 1-3 plus @pianorwc@.
Here are results using the @norm@ and @as-is@ methods, but with different sets of piano templates: first with three templates (1-3) and then with each template in turn as the only one.
The template turns out not to make an enormous difference -- perhaps because these recordings contain nothing but piano?
|Filename|@norm@/all|@as-is@/all|@norm@/3of4|@as-is@/3of4|@norm@/1|@as-is@/1|@norm@/2|@as-is@/2|@norm@/3|@as-is@/3|@norm@/rwc|@as-is@/rwc|
|@31.wav@|50|33|51|30|50|34|44|42|50|32|56|36|
|@MAPS_MUS-bach_846_AkPnBcht.wav@|87|15|86|16|86|24|75|20|73|10|71|18|
|@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|33|31|32|32|31|22|29|31|35|34|32|28|
|@MAPS_MUS-scn15_7_SptkBGAm.wav@|73|16|71|19|71|12|68|14|72|17|70|15|
|@mz_333_1MINp_align.wav@|66|3|68|1|63|4|67|2|67|1|63|3|
h4. For "generic" template set
The above results all use template sets with only piano templates in them.
Here are results using the @norm@ and @as-is@ methods, but with the full set of instrument templates (four pianos plus all the rest).
|Filename|@norm@|@as-is@|
|@31.wav@|49|37|
|@MAPS_MUS-bach_846_AkPnBcht.wav@|79|34|
|@MAPS_MUS-chpn_op7_1_ENSTDkAm.wav@|31|28|
|@MAPS_MUS-scn15_7_SptkBGAm.wav@|67|16|
|@mz_333_1MINp_align.wav@|63|5|
h4. Cross-checking with non-piano test data
The results need to be roughly comparable with those obtained from pre-normalised data using other datasets as well as the piano one. Here is a subset of the TRIOS dataset. The @norm@ result is that obtained from the plugin prior to doing this work, using pre-normalised data.
The @mirex@ result is that from the MIREX 2012 submission in MATLAB, but note that this always uses all instrument templates while the plugin results are based on selecting the "right" instrument for the piece (which is assumed to be the best, though we aren't actually testing that here).
|File|@mirex@|@norm@|@u4@|
|mozart/piano|60|64|56|
|mozart/viola|33|37|35|
|mozart/mix|51|58|55|
|mozart/clarinet|74|80|86|
|lussier/piano|45|52|63|
|lussier/mix|36|43|40|
|lussier/bassoon|43|75|80|
|lussier/trumpet|43|46|51|
|take_five/piano|61|46|69|
|take_five/mix|62|73|69|
|take_five/saxophone|78|80|84|