Refactor dependencies installation; Install via requirements.txt

.gitignore

@@ -0,0 +1,16 @@
+
+# Data
+data/
+
+# Virtualenv
+# http://iamzed.com/2009/05/07/a-primer-on-virtualenv/
+.Python
+[Bb]in
+[Ii]nclude
+[Ll]ib
+[Ll]ib64
+[Ll]ocal
+[Ss]cripts
+pyvenv.cfg
+.venv
+pip-selfcheck.json

README.md

@@ -1,5 +1,36 @@
 BPM Detector in Python
 =======================
+
+
+## Installation
+
+### Activate virtual environment
+``` bash
+source ./bin/activate
+```
+
+### Dependencies
+- numpy
+- PyWavelets
+- scipy
+- matplotlib
+``` bash
+pip install -r requirements.txt
+```
+
+## Test
+``` bash
+python ./bpm_detection/bpm_detection.py --filename ./data/*.wav 
+```
+
+## Help
+
+### Converting mp3
+``` bash
+mpg123 -w [path/to/original.mp3] [./data/wavfile.wav]
+```
+
+## Original Readme
 Implementation of a Beats Per Minute (BPM) detection algorithm, as presented in the paper of G. Tzanetakis, G. Essl and P. Cook titled: "Audio Analysis using the Discrete Wavelet Transform".
 
 You can find it here: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.63.5712
@@ -9,4 +40,3 @@ Based on the work done in the MATLAB code located at github.com/panagiop/the-BPM
 Process .wav file to determine the Beats Per Minute.
 
 Dependencies: scipy, numpy, pywt, matplotlib
-

bpm_detection/bpm_detection.py

@@ -12,7 +12,7 @@ def read_wav(filename):
     except IOError, e:
         print e
         return
-
+    # test eddit
     # typ = choose_type( wf.getsampwidth() ) #TODO: implement choose_type
     nsamps = wf.getnframes();
     assert(nsamps > 0);
@@ -50,9 +50,9 @@ def bpm_detector(data,fs):
     cD_sum = []
     levels = 4
     max_decimation = 2**(levels-1);
-    min_ndx = 60./ 220 * (fs/max_decimation)
-    max_ndx = 60./ 40 * (fs/max_decimation)
-    
+    min_ndx = int(60./ 220 * (fs/max_decimation))
+    max_ndx = int(60./ 40 * (fs/max_decimation))
+
     for loop in range(0,levels):
         cD = []
         # 1) DWT
@@ -90,14 +90,13 @@ def bpm_detector(data,fs):
     midpoint = len(correl) / 2
     correl_midpoint_tmp = correl[midpoint:]
     peak_ndx = peak_detect(correl_midpoint_tmp[min_ndx:max_ndx]);
+
     if len(peak_ndx) > 1:
         return no_audio_data()
 
     peak_ndx_adjusted = peak_ndx[0]+min_ndx;
     bpm = 60./ peak_ndx_adjusted * (fs/max_decimation)
-    print bpm
-    return bpm,correl
-
+    return bpm,correl,peak_ndx[0][0]
 
 if __name__ == '__main__':
     parser = argparse.ArgumentParser(description='Process .wav file to determine the Beats Per Minute.')
@@ -112,13 +111,14 @@ def bpm_detector(data,fs):
     data = []
     correl=[]
     bpm = 0
-    n=0;
+    #n=0;
     nsamps = len(samps)
-    window_samps = int(args.window*fs)         
-    samps_ndx = 0;  #first sample in window_ndx 
+    window_samps = int(args.window*fs)    # second * frame/second  = frames
+    samps_ndx = 0;  #first sample in window_ndx     # frames
     max_window_ndx = nsamps / window_samps;
     bpms = numpy.zeros(max_window_ndx)
 
+    last_bpm = None
     #iterate through all windows
     for window_ndx in xrange(0,max_window_ndx):
 
@@ -128,15 +128,23 @@ def bpm_detector(data,fs):
         if not ((len(data) % window_samps) == 0):
             raise AssertionError( str(len(data) ) ) 
 
-        bpm, correl_temp = bpm_detector(data,fs)
+        bpm, correl_temp, first_ndx = bpm_detector(data,fs)
         if bpm == None:
             continue
+
+        time_value = 1.0*(samps_ndx+first_ndx)/fs #frames / (frame/second) = second
+        bpm_value = bpm[0]
+        if last_bpm != bpm_value:
+            last_bpm = bpm_value
+
+            print "(Time, BPM): %.2f, %.2f"% (time_value, bpm_value)
+
         bpms[window_ndx] = bpm
         correl = correl_temp
 
         #iterate at the end of the loop
         samps_ndx = samps_ndx+window_samps;
-        n=n+1; #counter for debug...
+        #n=n+1; #counter for debug...
 
     bpm = numpy.median(bpms)
     print 'Completed.  Estimated Beats Per Minute:', bpm

requirements.txt

@@ -0,0 +1,4 @@
+numpy
+PyWavelets
+scipy
+matplotlib