Create app.py

app.py

@@ -0,0 +1,36 @@
+from transformers import pipeline
+import gradio as gr
+
+# Load the Whisper model
+whisper_pipe = pipeline("automatic-speech-recognition", model="openai/whisper-small")
+# Load the fine-tuned BERT model for harassment classification
+bert_pipe = pipeline("text-classification", model="abdulelahagr/harassment_lang_classifier")
+
+def classify_harassment(text):
+    predicted_category = bert_pipe(text)
+    return predicted_category
+
+
+def process_audio(speech_file):
+    whisper_result = whisper_pipe(speech_file, generate_kwargs={"language": "english"})
+    transcription = whisper_result["text"]
+
+    # 2. Classify the transcribed text
+    classification_result = classify_harassment(transcription)
+    predicted_label = classification_result[0]['label']
+    print(transcription, predicted_label)
+    # 3. Prepare results for display
+    return transcription, predicted_label
+
+
+
+with gr.Blocks() as demo:
+    gr.Markdown("## Kids harassment Classification")
+    audio_input = gr.Audio(type="filepath")
+    btn_process = gr.Button("Process")
+    transcription_output = gr.Textbox(label="Transcription")
+    classification_output = gr.Label(label="Classification Result")
+
+    btn_process.click(process_audio, inputs=audio_input, outputs=[transcription_output, classification_output])
+
+demo.launch(debug=True)