Update app.py

app.py

@@ -44,37 +44,39 @@ def classify_toxicity(audio_file, text_input, classify_anxiety):
     
     else:
         transcribed_text = text_input
-    print("classify anxiety ", classify_anxiety)
-    #### Toxicity Classifier ####
-        
-    toxicity_module = evaluate.load("toxicity",  "facebook/roberta-hate-speech-dynabench-r4-target")
-    #toxicity_module = evaluate.load("toxicity", 'DaNLP/da-electra-hatespeech-detection', module_type="measurement")
-
-    toxicity_results = toxicity_module.compute(predictions=[transcribed_text])
- 
-    toxicity_score = toxicity_results["toxicity"][0]
-    print(toxicity_score)
-
-    #### Text classification #####
-
-    device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
-
-    text_classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
-
-    sequence_to_classify = transcribed_text
-    print(classify_anxiety, class_options)
-    candidate_labels = class_options.get(classify_anxiety, [])
-    # classification_output = classifier(sequence_to_classify, candidate_labels, multi_label=False)
-    classification_output = text_classifier(sequence_to_classify, candidate_labels, multi_label=True)
-    print(classification_output)
-
-    #### Emotion classification ####
+    if classify_anxiety != "misophonia":
+        #### Toxicity Classifier ####
+            
+        toxicity_module = evaluate.load("toxicity",  "facebook/roberta-hate-speech-dynabench-r4-target")
+        #toxicity_module = evaluate.load("toxicity", 'DaNLP/da-electra-hatespeech-detection', module_type="measurement")
     
-    emotion_classifier = foreign_class(source="speechbrain/emotion-recognition-wav2vec2-IEMOCAP", pymodule_file="custom_interface.py", classname="CustomEncoderWav2vec2Classifier")
-    out_prob, score, index, text_lab = emotion_classifier.classify_file(audio_file)
- 
-    return toxicity_score, classification_output, emo_dict[text_lab[0]], transcribed_text
-    # return f"Toxicity Score ({available_models[selected_model]}): {toxicity_score:.4f}"
+        toxicity_results = toxicity_module.compute(predictions=[transcribed_text])
+     
+        toxicity_score = toxicity_results["toxicity"][0]
+        print(toxicity_score)
+    
+        #### Text classification #####
+    
+        device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
+    
+        text_classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
+    
+        sequence_to_classify = transcribed_text
+        print(classify_anxiety, class_options)
+        candidate_labels = class_options.get(classify_anxiety, [])
+        # classification_output = classifier(sequence_to_classify, candidate_labels, multi_label=False)
+        classification_output = text_classifier(sequence_to_classify, candidate_labels, multi_label=True)
+        print(classification_output)
+    
+        #### Emotion classification ####
+        
+        emotion_classifier = foreign_class(source="speechbrain/emotion-recognition-wav2vec2-IEMOCAP", pymodule_file="custom_interface.py", classname="CustomEncoderWav2vec2Classifier")
+        out_prob, score, index, text_lab = emotion_classifier.classify_file(audio_file)
+     
+        return toxicity_score, classification_output, emo_dict[text_lab[0]], transcribed_text
+        # return f"Toxicity Score ({available_models[selected_model]}): {toxicity_score:.4f}"
+    else: 
+        return classify_anxiety
  
 with gr.Blocks() as iface:
     with gr.Column():