How to do encode beats for many performances of one composition
Sun Jul  1 20:10:16 PDT 2007
Craig Stuart Sapp <craig@ccrma.stanford.edu>
=========================================================================

(1) Prepare note onset times in the directory:
     .../mazurka/website/ana/sapp-onset/
     by typing:
         make everything
     This will create onset files for all mazurkas which do not
     have onset files already.
     
(2) Download recordings for a particular mazurka.

(3) Convert the recording into WAVE files:
		for i in *.mp4
		do
		/cygdrive/d/faad/faad.exe $i
		echo $i is done
		done

(4) Create a listing of all files you will process
      ls pid*.wav > step1/filelist.txt
    Such as:
       	pid5667274b-06.wav
 	pid58917-06.wav
	pid9054b-12.wav
	pid9056-07.wav
	pid9057j-07.wav
	pid9059b-16.wav

(5) CD to step1 directory:
      cd step1

(6) Using the file dummy-basic.sv.out, run the program
       ./makesvfilesbasic dummy-basic.sv.out filelist.txt
    This will create SV project files for all files listed
    in filelist.txt.  Alternatively type:
       make
    And the SV files will be created and moved to the base directory.

(7) Run SV on all of the wave files using this command:
	for i in *.sv
	do
	/cygdrive/c/Program\ Files/Sonic\ Visualiser/sonic-visualiser.exe $i
	done

   Tap to each performance using the ';' key on the keyboard.
   Type control-S to save after finished tapping.

(8) Copy all of the saved .sv files into the step2 directory.

(9) Extract the tapping data to text files:
	./extracttaps *.sv

(10) Download onset data from
      http://mazurka.org.uk/auto/onset-sapp/mazurka*/pid*.onset010
      ./getonsets mazurka17-4 *.sv

(11) Snap all taps to nearest onsets:
	./snapall *.txt

(12)




===================================================

Creating Andrew input data files

(1) Save corrected tapping data as a Annotation layer (text file format)

(2) upload to website

(3) create directory mazurka/website/info/revcond/pid*

(4) go into that directory and type:
    ln -s ../Makefile.performance Makefile

(5) create measure labels for annotation layer data (called z.txt here)
       ../bin/makemeasurelabels 3 0 z.txt > pid*-clean.txt
    where 0 is the pickup count (1 = 1 pickup, 2 = 2 pickups)
    z.txt can then be deleted.

(6) Check that the measure count and pickup are correct in the -clean.txt file

(7) Add 9-line header in this format:

###performance-id:      pid9085-16
###title:               Mazurka in C-sharp minor, Op. 63, No. 3
###trial:               01/01
###date:                2007/03/28/
###reverse-conductor:   Craig Stuart Sapp
###performer:           Halina Czerny-Stefanska
###performance-date:    1949
###label:               Muza PNCD006
###label-title:         International Chopin Piano Competitions


(8) Create a tap file:
     ../bin/sv2revcond pid*-clean.txt > pid*-01.tap


(9) Create a kern score file (usually copying from another
    performer who plays the same piece and with the same
    repeat structure).

(10) type
       source /etc/profile.d/humdrum.sh
     to load Humdrum commands

(11) type:
	make allsingle


(12) go to parent directory and type:
         make index


===================================================


Store tempo changes in Excel Spreadsheet

(1) edit the file pid*-clean.txt and remove labels and header 
    (keeping only timings)

(2) copy and paste into Excel spreadsheet using existing 
    spreadsheets/worksheets as a guideline.