Create app.py

app.py

@@ -0,0 +1,51 @@
+import pandas as pd
+from pyannote.audio import Pipeline
+pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization",use_auth_token="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")
+def diarization():
+    diarization = pipeline("result.wav")
+    speakertime=[]
+# print the result
+    for turn, _, speaker in diarization.itertracks(yield_label=True):
+        details=[turn.start,turn.end,speaker]
+        speakertime.append(details)
+        #print(turn.start)
+        #print(speaker)
+        print(f"start={turn.start:.1f}s stop={turn.end:.1f}s speaker_{speaker}")
+    #print(speakertime)
+    df = pd.DataFrame(speakertime,columns=['start', 'end','speaker'])
+    text=[]
+    for i in range (df.start.count()):
+        text.append(generatetext("result.wav",df.start[i], df.end[i]))
+    df['text']=text
+    with open('my_file.txt', 'w') as my_file:
+      for i in range (df.start.count()):
+          my_file.write(df.speaker[i]+": " +df.text[i] + '\n')
+    print(open("my_file.txt","r").read())
+
+
+def generatetext(filename,starttime,endtime):
+  
+  t1 = starttime * 1000 # works in milliseconds
+  t2 = endtime * 1000
+
+  newAudio = AudioSegment.from_wav(filename)
+  a = newAudio[t1:t2]
+  a.export('audio.wav', format="wav") 
+  text1 = whisper('audio.wav')
+  return text1.get("text")
+
+block = gr.Blocks()
+with block:
+    with gr.Group():
+        with gr.Box(): 
+            with gr.Row().style():
+               
+                inp_audio = gr.Audio(
+                    label="Input Audio",
+                    type="filepath",
+                    mirror_webcam = False
+                )
+                outputdialogs = gr.Textbox()
+            btn = gr.Button("Generate Text")
+        btn.click(diarisation, inputs=[inp_audio], outputdialogs=[op],api_name="view_api")
+block.launch(enable_queue = True,debug=True)