updated: prosodic features

Presently/presently/users/emotiondetectionvideo.py

-# -*- coding: utf-8 -*-
-"""emotionDetection.ipynb
-
-Automatically generated by Colaboratory.
-
-Original file is located at
-    https://colab.research.google.com/drive/1o8opa083KfLohUfWs-oGfkdo9kuRN_eS
-
-**Importing dependencies**
-
-
-*   pip install fer
-*   List item
-"""
-
 from fer import Video
 from fer import FER
 import os
@@ -20,13 +5,9 @@ import sys
 import pandas as pd
 import numpy as np
 
-"""Upload the video to the Google colab and get the path of the video and xml file"""
-
 location_videofile = "F:\\CDAP-PRESENTLY\\21_22-j-02\\Presently\\presently\\media\\video\\22\\testvideo.mp4"
 cascPath = "F:\\CDAP-PRESENTLY\\21_22-j-02\\Presently\\presently\\users\\models\\abc.xml"
 
-"""Get labels to each emotion"""
-
 def _get_labels():
   return {
             0: "angry",
@@ -38,10 +19,7 @@ def _get_labels():
             6: "neutral",
         }
 
-"""Build a square (around the face)"""
-
 def tosquare(bbox):
-        """Convert bounding box to square by elongating shorter side."""
         x, y, w, h = bbox
         if h > w:
             diff = h - w
@@ -56,15 +34,11 @@ def tosquare(bbox):
 
         return (x, y, w, h)
 
-"""Builded square around the face"""
-
 def __apply_offsets(face_coordinates):
   x, y, width, height = face_coordinates
   x_off, y_off = (10, 10)
   return (x - x_off, x + width + x_off, y - y_off, y + height + y_off)
 
-"""To correct numbers"""
-
 def __preprocess_input(x, v2=False):
         x = x.astype("float32")
         x = x / 255.0
@@ -73,8 +47,6 @@ def __preprocess_input(x, v2=False):
             x = x * 2.0
         return x
 
-"""Add padding to the image"""
-
 def pad(image):
         PADDING = 40
         row, col = image.shape[:2]
@@ -106,7 +78,6 @@ def detect_emotions(location_videofile, NumberofFrames):
 
   success,image = vidcap.read()
   frame_count = vidcap.get(cv2.CAP_PROP_FRAME_COUNT)
-  print("Frame Count: ", frame_count)
   count = 0
   faceCascade = cv2.CascadeClassifier(cascPath)
 
@@ -116,7 +87,7 @@ def detect_emotions(location_videofile, NumberofFrames):
 
       if success:
           if frame_count > NumberofFrames+1:
-            count += frame_count/(NumberofFrames+1) # i.e. at 30 fps, this advances one second
+            count += frame_count/(NumberofFrames+1) 
           else:
             count += 1
           vidcap.set(cv2.CAP_PROP_POS_FRAMES, count)
@@ -135,7 +106,6 @@ def detect_emotions(location_videofile, NumberofFrames):
                 face_coordinates = tosquare(face_coordinates)
                 x1, x2, y1, y2 = __apply_offsets(face_coordinates)
                 
-                # adjust for padding
                 x1 += PADDING
                 x2 += PADDING
                 y1 += PADDING
@@ -154,10 +124,8 @@ def detect_emotions(location_videofile, NumberofFrames):
                 try:
                   gray_face = cv2.resize(gray_face, model.input_shape[1:3])
                 except Exception as e:
-                  #print("Cannot resize")
                   continue
                 
-                # Local Keras model
                 gray_face = __preprocess_input(gray_face, True)
                 gray_face = np.expand_dims(np.expand_dims(gray_face, 0), -1)
 
@@ -170,9 +138,6 @@ def detect_emotions(location_videofile, NumberofFrames):
                 emotions.append(
                     dict(box=face_coordinates, emotions=labelled_emotions)
                 )
-            #print("Prediction : ", emotions[0]["emotions"])
-            #plt.imshow(gray_img, interpolation='nearest')
-            #plt.show()
             top_emotions  = [max(e["emotions"], key=lambda key: e["emotions"][key]) for e in emotions]
             if len(top_emotions):
               for top_emotion in emotions[0]["emotions"]:
@@ -190,4 +155,3 @@ def detect_emotions(location_videofile, NumberofFrames):
     return max(arry, key=arry.get), arry
 
 emo, arr1 = detect_emotions(location_videofile, 300)
-# print(emo, arr1)
\ No newline at end of file

Presently/presently/users/myprosody.py

@@ -25,8 +25,8 @@ def run_praat_file(m, p):
 
     try:
         objects= run_file(sourcerun, -20, 2, 0.3, "yes",sound,path, 80, 400, 0.01, capture_output=True)
-        print (objects[0]) # This will print the info from the sound object, and objects[0] is a parselmouth.Sound object
-        z1=str( objects[1]) # This will print the info from the textgrid object, and objects[1] is a parselmouth.Data object with a TextGrid inside
+        print (objects[0])
+        z1=str( objects[1])
         z2=z1.strip().split()
         return z2
     except:
@@ -39,10 +39,9 @@ def myspsyl(m,p):
     Detect and count number of syllables
     """
     z2 = run_praat_file(m, p)
-    z3=int(z2[0]) # will be the integer number 10
-    z4=float(z2[3]) # will be the floating point number 8.3
-    print ("number_ of_syllables=",z3)
-    return z3
+    z3=int(z2[0])
+    z4=float(z2[3])
+    return ("Number of syllables = ",z3)
 
 def mysppaus(m,p):
     """
@@ -51,8 +50,7 @@ def mysppaus(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[1]) # will be the integer number 10
     z4=float(z2[3]) # will be the floating point number 8.3
-    print ("number_of_pauses=",z3)
-    return z3
+    return ("Number of pauses = ",z3)
 
 def myspsr(m,p):
     """
@@ -61,8 +59,7 @@ def myspsr(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[2]) # will be the integer number 10
     z4=float(z2[3]) # will be the floating point number 8.3
-    print ("rate_of_speech=",z3,"# syllables/sec original duration")
-    return z3
+    return ("Rate of speech = ",z3," # syllables/sec original duration")
 
 def myspatc(m,p):
     """
@@ -71,8 +68,7 @@ def myspatc(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[3]) # will be the integer number 10
     z4=float(z2[3]) # will be the floating point number 8.3
-    print ("articulation_rate=",z3,"# syllables/sec speaking duration")
-    return z3
+    return ("Articulation Rate =",z3," # syllables/sec speaking duration")
 
 def myspst(m,p):
     """
@@ -81,8 +77,7 @@ def myspst(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[3]) # will be the integer number 10
     z4=float(z2[4]) # will be the floating point number 8.3
-    print ("speaking_duration=",z4,"# sec only speaking duration without pauses")
-    return z4
+    return ("Speaking duration = ",z4," # sec only speaking duration without pauses")
 
 def myspod(m,p):
     """
@@ -91,8 +86,7 @@ def myspod(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[3]) # will be the integer number 10
     z4=float(z2[5]) # will be the floating point number 8.3
-    print ("original_duration=",z4,"# sec total speaking duration with pauses")
-    return z4
+    return ("Original duration = ",z4," # sec total speaking duration with pauses")
 
 def myspbala(m,p):
     """
@@ -101,8 +95,7 @@ def myspbala(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[3]) # will be the integer number 10
     z4=float(z2[6]) # will be the floating point number 8.3
-    print ("balance=",z4,"# ratio (speaking duration)/(original duration)")
-    return z4
+    return ("Balance = ",z4," # ratio (speaking duration)/(original duration)")
 
 def myspf0mean(m,p):
     """
@@ -111,8 +104,7 @@ def myspf0mean(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[3]) # will be the integer number 10
     z4=float(z2[7]) # will be the floating point number 8.3
-    print ("f0_mean=",z4,"# Hz global mean of fundamental frequency distribution")
-    return z4
+    return ("f0 mean = ",z4," # Hz global mean of fundamental frequency distribution")
 
 def myspf0sd(m,p):
     """
@@ -121,8 +113,7 @@ def myspf0sd(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[3]) # will be the integer number 10
     z4=float(z2[8]) # will be the floating point number 8.3
-    print ("f0_SD=",z4,"# Hz global standard deviation of fundamental frequency distribution")
-    return z4
+    return ("f0 SD = ",z4," # Hz global standard deviation of fundamental frequency distribution")
 
 def myspf0med(m,p):
     """
@@ -131,8 +122,7 @@ def myspf0med(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[3]) # will be the integer number 10
     z4=float(z2[9]) # will be the floating point number 8.3
-    print ("f0_MD=",z4,"# Hz global median of fundamental frequency distribution")
-    return z4
+    return ("f0 MD = ",z4," # Hz global median of fundamental frequency distribution")
 
 def myspf0min(m,p):
     """
@@ -141,8 +131,7 @@ def myspf0min(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[10]) # will be the integer number 10
     z4=float(z2[10]) # will be the floating point number 8.3
-    print ("f0_min=",z3,"# Hz global minimum of fundamental frequency distribution")
-    return z3
+    return ("f0 min = ",z3," # Hz global minimum of fundamental frequency distribution")
 
 def myspf0max(m,p):
     """
@@ -151,8 +140,7 @@ def myspf0max(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[11]) # will be the integer number 10
     z4=float(z2[11]) # will be the floating point number 8.3
-    print ("f0_max=",z3,"# Hz global maximum of fundamental frequency distribution")
-    return z3
+    return ("f0 max = ",z3," # Hz global maximum of fundamental frequency distribution")
 
 def myspf0q25(m,p):
     """
@@ -161,8 +149,7 @@ def myspf0q25(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[12]) # will be the integer number 10
     z4=float(z2[11]) # will be the floating point number 8.3
-    print ("f0_quan25=",z3,"# Hz global 25th quantile of fundamental frequency distribution")
-    return z3
+    return ("f0 quan25 = ",z3," # Hz global 25th quantile of fundamental frequency distribution")
 
 def myspf0q75(m,p):
     """
@@ -171,22 +158,7 @@ def myspf0q75(m,p):
     z2 = run_praat_file(m, p)
     z3=int(z2[13]) # will be the integer number 10
     z4=float(z2[11]) # will be the floating point number 8.3
-    print ("f0_quan75=",z3,"# Hz global 75th quantile of fundamental frequency distribution")
-    return z3
-
-def mysptotal(m,p):
-    """
-    Overview
-    """
-    z2 = run_praat_file(m, p)
-    z3=np.array(z2)
-    z4=np.array(z3)[np.newaxis]
-    z5=z4.T
-    dataset=pd.DataFrame({"number_ of_syllables":z5[0,:],"number_of_pauses":z5[1,:],"rate_of_speech":z5[2,:],"articulation_rate":z5[3,:],"speaking_duration":z5[4,:],
-                        "original_duration":z5[5,:],"balance":z5[6,:],"f0_mean":z5[7,:],"f0_std":z5[8,:],"f0_median":z5[9,:],"f0_min":z5[10,:],"f0_max":z5[11,:],
-                        "f0_quantile25":z5[12,:],"f0_quan75":z5[13,:]})
-    print (dataset.T)
-    return dataset.T
+    return ("f0 quan75 = ",z3," # Hz global 75th quantile of fundamental frequency distribution")
 
 def mysppron(m,p):
     """
@@ -206,9 +178,9 @@ def mysppron(m,p):
         db= binom.rvs(n=10,p=z4,size=10000)
         a=np.array(db)
         b=np.mean(a)*100/10
-        print ("Pronunciation_posteriori_probability_score_percentage= :%.2f" % (b))
+        return ("Pronunciation posteriori probability score percentage = : %.2f" % (b))
     except:
-        print ("Try again the sound of the audio was not clear")
+        return ("Try again the sound of the audio was not clear")
     return
 
 def myspgend(m,p):
@@ -266,21 +238,21 @@ def myspgend(m,p):
         else:
             mmm=0.35
         if z4>97 and z4<=114:
-            print("a Male, mood of speech: Showing no emotion, normal, p-value/sample size= :%.2f" % (mmm), (nnn))
+            return ("A Male, Mood of speech: Showing no emotion | normal, p-value/sample size :%.2f" % (mmm), (nnn))
         elif z4>114 and z4<=135:
-            print("a Male, mood of speech: Reading, p-value/sample size= :%.2f" % (mmm), (nnn))
+            return ("A Male, Mood of speech: Reading, p-value/sample size :%.2f" % (mmm), (nnn)) 
         elif z4>135 and z4<=163:
-            print("a Male, mood of speech: speaking passionately, p-value/sample size= :%.2f" % (mmm), (nnn))
+            return ("A Male, Mood of speech: speaking passionately, p-value/sample size :%.2f" % (mmm), (nnn))
         elif z4>163 and z4<=197:
-            print("a female, mood of speech: Showing no emotion, normal, p-value/sample size= :%.2f" % (mmm), (nnn))
+            return ("A Female, Mood of speech: Showing no emotion | normal, p-value/sample size :%.2f" % (mmm), (nnn))
         elif z4>197 and z4<=226:
-            print("a female, mood of speech: Reading, p-value/sample size= :%.2f" % (mmm), (nnn))
+            return ("A Female, Mood of speech: Reading, p-value/sample size :%.2f" % (mmm), (nnn))
         elif z4>226 and z4<=245:
-            print("a female, mood of speech: speaking passionately, p-value/sample size= :%.2f" % (mmm), (nnn))
+            return ("A Female, Mood of speech: speaking passionately, p-value/sample size :%.2f" % (mmm), (nnn))
         else:
-            print("Voice not recognized")
+            return ("Voice not recognized")
     except:
-        print ("Try again the sound of the audio was not clear")
+        return  ("Try again the sound of the audio was not clear")
 
 def myprosody(m,p):
     """
@@ -299,11 +271,10 @@ def myprosody(m,p):
     result_array = np.empty((0, 27))
     try:
         objects= run_file(sourcerun, -20, 2, 0.3, "yes",sound,path, 80, 400, 0.01, capture_output=True)
-        z1=( objects[1]) # This will print the info from the textgrid object, and objects[1] is a parselmouth.Data object with a TextGrid inside
+        z1=( objects[1])
         z3=z1.strip().split()
         z2=np.array([z3])
         result_array=np.append(result_array,[z3], axis=0)
-        #print(z3)
         np.savetxt(outo,result_array, fmt='%s',delimiter=',')
         #Data and features analysis
         df = pd.read_csv(outo,
@@ -334,15 +305,15 @@ def myprosody(m,p):
             if he==0:
                 he=25
                 dfout = "%s:\t %f (%s)" %  (nsns[i],he,"% percentile ")
-                print(dfout)
+                return(dfout)
             elif he>=25 and he<=75:
                 dfout = "%s:\t %f (%s)" % (nsns[i],he,"% percentile ")
-                print(dfout)
+                return(dfout)
             else:
                 dfout = "%s:\t (%s)" % (nsns[i],":Out of Range")
-                print(dfout)
+                return(dfout)
     except:
-        print ("Try again the sound of the audio was not clear")	
+        return ("Try again the sound of the audio was not clear")	
     
 def mysplev(m,p):
 	import sys

Presently/presently/users/prosodydetectionaudio.py

-import myprosody as mysp
-import pickle
-
-p="happy" 
-c="F:\\CDAP-PRESENTLY\\21_22-j-02\\Presently\\presently\\users\\myprosody" 
-
-mysp.myspsyl(p,c)
-mysp.mysppaus(p,c)
-mysp.myspsr(p,c)
-mysp.myspatc(p,c)
-mysp.myspst(p,c)
-mysp.myspod(p,c)
-mysp.myspbala(p,c)
-mysp.myspf0mean(p,c)
-mysp.myspf0sd(p,c)
-mysp.myspf0med(p,c)
-mysp.myspf0min(p,c)
-mysp.myspf0max(p,c)
-mysp.myspf0q25(p,c)
-mysp.myspf0q75(p,c)
-
-mysp.mysptotal(p,c)
-mysp.myspgend(p,c)
-mysp.mysppron(p,c)
-mysp.myprosody(p,c)
-mysp.mysplev(p,c)

Presently/presently/users/templates/users/base.html

@@ -7,8 +7,7 @@
     <link rel="stylesheet" href="{% static 'users/css/main.css' %}" type="text/css">
     <link rel="stylesheet" href="{% static 'users/css/plugin.css' %}" type="text/css">
     <link rel="stylesheet" href="{% static 'users/css/style.css' %}" type="text/css">
-
-
+    <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/4.7.0/css/font-awesome.min.css">
     <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.0.0-beta3/dist/css/bootstrap.min.css" rel="stylesheet" integrity="sha384-eOJMYsd53ii+scO/bJGFsiCZc+5NDVN2yr8+0RDqr0Ql0h+rP48ckxlpbzKgwra6" crossorigin="anonymous">
     <link rel="preconnect" href="https://fonts.gstatic.com">
     <link href="https://fonts.googleapis.com/css2?family=Montserrat:wght@500&family=Open+Sans:wght@800&display=swap" rel="stylesheet">

Presently/presently/users/templates/users/test.html

@@ -5,7 +5,7 @@
 {% block content %}
     <h2> {{ var1 }} {{ var2 }}</h2>
 
-    <!-- {% for key, value in predictions.items %}
+    <!-- {% for key, value in var2.items %}
             <h2>{{ forloop.counter }}. {{ key }} : {{ value }}</h2>
         {% endfor %} -->
     

Presently/presently/users/views.py

@@ -10,11 +10,16 @@ from .models import Video
 from django.template import Template, Context
 import datetime
 from . import emotiondetectionvideo
-from . import emotiondetectionaudio
+# from . import emotiondetectionaudio
+import myprosody as mysp
+import pickle
+from . import myprosody
 
 def test(request):
-    var1= emotiondetectionaudio.emo_list
-    var2= emotiondetectionaudio.total_predictions_np
+    p="happy" 
+    c="F:\\CDAP-PRESENTLY\\21_22-j-02\\Presently\\presently\\users\\myprosody" 
+    var2= myprosody.myspgend(p,c)
+    var1= myprosody.myprosody(p,c)
     context= {
         'var1': var1,
         'var2': var2
@@ -60,7 +65,46 @@ def emotionvideo(request):
     return render(request, 'users/emotionvideo.html',context) 
 
 def emotionaudioprosody(request):
-    return render(request, 'users/prosody.html')    
+    p="happy" 
+    c="F:\\CDAP-PRESENTLY\\21_22-j-02\\Presently\\presently\\users\\myprosody" 
+    myspsyl= myprosody.myspsyl(p,c)
+    mysppaus= myprosody.mysppaus(p,c)
+    myspsr= myprosody.myspsr(p,c)
+    myspatc= myprosody.myspatc(p,c)
+    myspst= myprosody.myspst(p,c)
+    myspod= myprosody.myspod(p,c)
+    myspbala= myprosody.myspbala(p,c)
+    myspf0mean= myprosody.myspf0mean(p,c)
+    myspf0sd= myprosody.myspf0sd(p,c)
+    myspf0med= myprosody.myspf0med(p,c)
+    myspf0min= myprosody.myspf0min(p,c)
+    myspf0max= myprosody.myspf0max(p,c)
+    myspf0q25= myprosody.myspf0q25(p,c)
+    myspf0q75= myprosody.myspf0q75(p,c)
+    myspgend= myprosody.myspgend(p,c)
+    mysppron= myprosody.mysppron(p,c)
+    prosody= myprosody.myprosody(p,c)
+    
+    context= {
+        'myspsyl': myspsyl,
+        'mysppaus': mysppaus,
+        'myspsr': myspsr,
+        'myspatc': myspatc,
+        'myspst': myspst,
+        'myspod': myspod,
+        'myspbala': myspbala,
+        'myspf0mean': myspf0mean,
+        'myspf0sd': myspf0sd,
+        'myspf0med': myspf0med,
+        'myspf0min': myspf0min,
+        'myspf0max': myspf0max,
+        'myspf0q25': myspf0q25,
+        'myspf0q75': myspf0q75,
+        'myspgend': myspgend,
+        'mysppron': mysppron,
+        'prosody': prosody
+    }
+    return render(request, 'users/prosody.html',context)    
 
 def overallfeedback(request):
     return render(request, 'users/overallfeedback.html')