updated requirements.txt

app.py

@@ -1,43 +1,40 @@
 import gradio as gr
-from transformers import pipeline, BertTokenizer, EncoderDecoderModel
+from transformers import BertTokenizer, EncoderDecoderModel
 import torch
 
-# Load Whisper for speech-to-text
-asr = pipeline("automatic-speech-recognition", model="openai/whisper-small")
+# Load model and tokenizer
+model = EncoderDecoderModel.from_pretrained("Imsachinsingh00/bert2bert-mts-summary")
+tokenizer = BertTokenizer.from_pretrained("Imsachinsingh00/bert2bert-mts-summary")
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model = model.to(device)
 
-# Load your fine-tuned summarization model
-model_name = "Imsachinsingh00/bert2bert-mts-summary"
-tokenizer = BertTokenizer.from_pretrained(model_name)
-model = EncoderDecoderModel.from_pretrained(model_name).to("cuda" if torch.cuda.is_available() else "cpu")
-
-# Summarization function
-def summarize_text(text):
-    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=512).to(model.device)
+def summarize_text(dialogue):
+    inputs = tokenizer(dialogue, return_tensors="pt", padding=True, truncation=True, max_length=512)
+    inputs = {k: v.to(device) for k, v in inputs.items()}
     outputs = model.generate(**inputs, max_length=64)
     summary = tokenizer.decode(outputs[0], skip_special_tokens=True)
     return summary
 
-# Pipeline: audio → transcription
-def transcribe(audio):
-    return asr(audio)["text"]
-
-# App UI
+# Gradio app
 with gr.Blocks() as demo:
-    gr.Markdown("## 🩺 Medical Dialogue Summarizer")
+    gr.Markdown("## 🎤 Medical Dialogue Summarization App")
 
     with gr.Row():
-        with gr.Column():
-            audio_input = gr.Audio(source="microphone", type="filepath", label="🎙️ Record Dialogue")
-            transcribed_text = gr.Textbox(lines=10, label="📝 Transcribed Text (editable)")
-
-            record_button = gr.Button("🎧 Transcribe")
-            record_button.click(transcribe, inputs=audio_input, outputs=transcribed_text)
-
-        with gr.Column():
-            summary_output = gr.Textbox(lines=10, label="📋 Summary (output)", interactive=False)
-            summarize_button = gr.Button("✂️ Summarize")
-            summarize_button.click(summarize_text, inputs=transcribed_text, outputs=summary_output)
-
-    gr.Markdown("Built for Voize Interview — Powered by Whisper + BERT")
+        mic_input = gr.Microphone(label="🎙️ Record Dialogue")
+        text_input = gr.Textbox(label="📝 Or Paste Dialogue", lines=10, placeholder="Paste or speak a conversation here...")
+
+    summarize_btn = gr.Button("Summarize")
+    summary_output = gr.Textbox(label="📄 Summary", interactive=False)
+
+    def transcribe_and_summarize(audio, text):
+        import whisper
+        model_whisper = whisper.load_model("base")
+        result = model_whisper.transcribe(audio)
+        transcribed_text = result["text"]
+        text = text or transcribed_text
+        summary = summarize_text(text)
+        return transcribed_text, summary
+
+    summarize_btn.click(fn=transcribe_and_summarize, inputs=[mic_input, text_input], outputs=[text_input, summary_output])
 
 demo.launch()

requirements.txt

@@ -5,4 +5,5 @@ rouge_score
 torch
 gradio
 scikit-learn
-huggingface_hub
+huggingface_hub
+whisper