Update app.py

app.py

@@ -47,8 +47,8 @@ def createlex(filename):
             data_dict[key] = value
     return data_dict
 
-lex=createlex("num_words_tel.txt")   
-
+tellex=createlex("num_words_tel.txt")   
+kanlex=createlex("num_words_kn.txt")
 def addnum(inlist):
     sum=0
     for num in inlist:
@@ -76,7 +76,7 @@ def get_val(word, lexicon):
             return None
     else:
         return None
-def convert2num(input, lex):
+def convert2numtel(input, lex):
     input += " #"  # Add a period for termination
     words = input.split()
     i = 0
@@ -131,6 +131,81 @@ def convert2num(input, lex):
     # Final processing
     outstr = outstr.replace('#','')  # Remove trailing spaces
     return outstr
+
+def convert2numkn(input, lex):
+    input += " ######"  # Add a period for termination
+    words = input.split()
+    i = 0
+    num = 0
+    outstr = ""
+    digit_end = True
+    numlist = []
+    addflag = False
+    
+    prevword = []
+    
+    # Process the words
+    while i < len(words):
+        
+        # Get numerical value of the word
+        numval = get_val(words[i], lex)
+        if len(prevword)>=3:
+            prevword.pop(0)
+            prevword.append(words[i])
+        else:
+            prevword.append(words[i])
+        if numval is not None:
+           
+            addflag = True
+           
+            numlist.append(numval)
+           
+        else:
+           
+            
+            #print("word--->"+words[i])
+            #print("addflagword--->"+str(addflag))
+            prevwords=" ".join(prevword)
+            #print("prev word--->"+prevwords)
+            numval=get_val(prevwords,lex)
+            if numval is not None:
+                #addflag=True
+                #print("numval " +numval)
+                numlist=[]
+                #print("First outstr--->"+outstr)
+                
+                
+                outwords = outstr.split()
+                outstr=" ".join(outwords[:-1])
+                #print("outstr--->"+outstr)
+                
+                outstr += " " + str(numval) + " " 
+                #print(" aoutstr--->"+outstr)
+                numval=0
+                addflag=False
+                
+            else:
+                if addflag:
+                    num = addnum(numlist)
+                    outstr += str(num) + " " + words[i] + " "
+                    #print("penlast outstr--->"+outstr)
+                    addflag = False
+                    numlist = []
+                else:
+                    outstr += words[i] + " "
+                    #print("last outstr--->"+outstr)
+                if not digit_end:
+                    digit_end = True
+        
+        
+       
+        # Move to the next word
+        i += 1
+
+    # Final processing
+    outstr = outstr.replace('#','')  # Remove trailing spaces
+    return outstr
+    
 # Function to apply a high-pass filter
 def high_pass_filter(audio, sr, cutoff=300):
     nyquist = 0.5 * sr
@@ -163,7 +238,7 @@ def recognize_speech_telugu(audio_file):
     text_value = result['text']
     print (text_value)
     cleaned_text = text_value.replace("<s>", "")
-    converted_text=convert2num(cleaned_text,lex)
+    converted_text=convert2numtel(cleaned_text,tellex)
     # cleaned_text=convert2num(cleaned_text,lex)
     # converted_to_list = convert_to_list(cleaned_text, text_to_list())
     # processed_doubles = process_doubles(converted_to_list)
@@ -181,6 +256,7 @@ def recognize_speech_kannada(audio_file):
     result = asr_model_kannada(denoised_audio)
     text_value = result['text']
     cleaned_text = text_value.replace("[UNK]", "")
+    converted_text=convert2numkn(cleaned_text,kanlex)
     #converted_text=convert2num(cleaned_text,lex)
     # cleaned_text=convert2num(cleaned_text,lex)
     # converted_to_list = convert_to_list(cleaned_text, text_to_list())