Update app.py

app.py

@@ -34,16 +34,15 @@ def classifier(word_list_with_timestamp, language):
     else:
         list_to_search = set(english_bad_Words["words"])
     
-    output = []
+    foul_words = []
     negative_timestamps = []
     for item in word_list_with_timestamp:
         word = clean_text(item['text'])
         if word in list_to_search:
-            output.append((item['text'], "negative"))
+            if word not in foul_words:
+                foul_words.append(word)
             negative_timestamps.append(item['timestamp'])
-        else:
-            output.append((item['text'], "positive"))
-    return [output, negative_timestamps]
+    return [foul_words, negative_timestamps]
 
 def generate_bleep(duration_ms, frequency=1000):
     sine_wave = Sine(frequency)
@@ -86,13 +85,16 @@ def transcribe(input_audio, audio_language, task, timestamp_type):
 
     timestamps = format_output_to_list(output['chunks'])
 
-    classifier_output, negative_timestamps = classifier(output['chunks'], audio_language)
+    foul_words_list, negative_timestamps = classifier(output['chunks'], audio_language)
 
+    foul_words_list = ", ".join(foul_words_list)
+    
+    
     audio_output = mute_audio_range(input_audio, negative_timestamps)
     output_path = "output_audio.wav"
     audio_output.export(output_path, format="wav")
 
-    return  [text, timestamps, classifier_output, output_path]
+    return  [text, timestamps, foul_words_list, output_path]
 
 examples = [
         ["arabic_english_audios/audios/arabic_audio_1.wav", 'Arabic', 'transcribe', 'word'],
@@ -126,10 +128,10 @@ with gr.Blocks(theme=gr.themes.Default()) as demo:
         with gr.Column():
             transcript_output = gr.Text(label="Transcript")
             timestamp_output = gr.Text(label="Timestamps")
-            highlighted_output = gr.HighlightedText(label="Words Classification", combine_adjacent=True, show_legend=True, color_map={"negative": "red", "positive": "green"})
+            foul_words_output = gr.Text(label="Foul words in Audio")
             output_audio = gr.Audio(label="Output Audio")
     
-    examples = gr.Examples(examples, inputs=[audio_input, audio_language, task, timestamp_type], outputs=[transcript_output, timestamp_output], fn=transcribe, examples_per_page=20)
+    examples = gr.Examples(examples, inputs=[audio_input, audio_language, task, timestamp_type], outputs=[transcript_output, timestamp_output, foul_words_output, output_audio], fn=transcribe, examples_per_page=20)
 
     submit_button.click(fn=transcribe, inputs=[audio_input, audio_language, task, timestamp_type], outputs=[transcript_output, timestamp_output, highlighted_output, output_audio])
     clear_button.add([audio_input, audio_language, task, timestamp_type, transcript_output, timestamp_output, highlighted_output, output_audio])