Update app.py

app.py

@@ -72,12 +72,17 @@ def transcribe(
     # and other debugging detailed element you might be
     # interested in
     
+    
     segments = response.json()["prediction"]
 
     output = ""
-
+    current_speaker = ""
     for segment in segments:
-        output = output + " " + segment["transcription"]
+        if segment["speaker"] != current_speaker and segment["speaker"]!= "unknown":
+            current_speaker = segment["speaker"]
+            output = output + "<br/><br/><b>" + segment["speaker"] + ":</b> " + segment["transcription"]
+        else:
+            output = output + " " + segment["transcription"]
     
 
     return output, response.json()["prediction_raw"]
@@ -97,7 +102,7 @@ iface = gr.Interface(
     inputs=[
         gr.Audio(label="Audio file", source="upload", type="filepath"),
     ],
-    outputs=["text", "json"]
+    outputs=["html", "json"],
     examples=[
         ["examples/good.will.hunting.wav"],
         ["examples/wolf.of.wall.street.wav"],