Upload gradio_app.py

gradio_app.py

@@ -0,0 +1,302 @@
+# # # if you dont use pipenv uncomment the following:
+# # # from dotenv import load_dotenv
+# # # load_dotenv()
+
+# # #VoiceBot UI with Gradio
+# # import os
+# # import gradio as gr
+
+# # from brain_of_the_doctor import encode_image, analyze_image_with_query
+# # from voice_of_the_patient import record_audio, transcribe_with_groq
+# # from voice_of_the_doctor import text_to_speech_with_gtts, text_to_speech_with_elevenlabs
+
+# # #load_dotenv()
+
+# # system_prompt="""You have to act as a professional doctor, i know you are not but this is for learning purpose. 
+# #             What's in this image?. Do you find anything wrong with it medically? 
+# #             If you make a differential, suggest some remedies for them. Donot add any numbers or special characters in 
+# #             your response. Your response should be in one long paragraph. Also always answer as if you are answering to a real person.
+# #             Donot say 'In the image I see' but say 'With what I see, I think you have ....'
+# #             Dont respond as an AI model in markdown, your answer should mimic that of an actual doctor not an AI bot, 
+# #             Keep your answer concise (max 2 sentences). No preamble, start your answer right away please"""
+
+
+# # def process_inputs(audio_filepath, image_filepath):
+# #     speech_to_text_output = transcribe_with_groq(GROQ_API_KEY=os.environ.get("GROQ_API_KEY"), 
+# #                                                  audio_filepath=audio_filepath,
+# #                                                  stt_model="whisper-large-v3")
+
+# #     # Handle the image input
+# #     if image_filepath:
+# #         doctor_response = analyze_image_with_query(query=system_prompt+speech_to_text_output, encoded_image=encode_image(image_filepath), model="meta-llama/llama-4-scout-17b-16e-instruct") #model="meta-llama/llama-4-maverick-17b-128e-instruct") 
+# #     else:
+# #         doctor_response = "No image provided for me to analyze"
+
+# #     # voice_of_doctor = text_to_speech_with_elevenlabs(input_text=doctor_response, output_filepath="final.mp3") 
+# #     voice_of_doctor = text_to_speech_with_gtts(input_text=doctor_response, output_filepath="final.mp3") 
+
+# #     return speech_to_text_output, doctor_response, voice_of_doctor
+
+
+# # # Create the interface
+# # iface = gr.Interface(
+# #     fn=process_inputs,
+# #     inputs=[
+# #         gr.Audio(sources=["microphone"], type="filepath"),
+# #         gr.Image(type="filepath")
+# #     ],
+# #     outputs=[
+# #         gr.Textbox(label="Speech to Text"),
+# #         gr.Textbox(label="Doctor's Response"),
+# #         gr.Audio("Temp.mp3")
+# #     ],
+# #     title="AI Doctor with Vision and Voice"
+# # )
+
+# # iface.launch(debug=True)
+
+# # #http://127.0.0.1:7860
+
+
+
+# # if you dont use pipenv uncomment the following:
+# # from dotenv import load_dotenv
+# # load_dotenv()
+
+
+
+
+
+# # ---------------------------------------------------------
+
+
+
+
+# # VoiceBot UI with Gradio
+# import os
+# import gradio as gr
+
+# from brain_of_the_doctor import encode_image, analyze_image_with_query
+# from voice_of_the_patient import record_audio, transcribe_with_groq
+# from voice_of_the_doctor import text_to_speech_with_gtts, text_to_speech_with_elevenlabs
+
+# # load_dotenv()
+
+# system_prompt = """
+# You have to act as a professional doctor, i know you are not but this is for learning purpose. 
+# What's in this image? Do you find anything wrong with it medically? 
+# If you make a differential, suggest some remedies for them. Donot add any numbers or special characters in your response. 
+# Your response should be in one long paragraph. Also always answer as if you are answering to a real person.
+# Donot say 'In the image I see' but say 'With what I see, I think you have ....'
+# Dont respond as an AI model in markdown, your answer should mimic that of an actual doctor not an AI bot, 
+# Keep your answer concise (max 2 sentences). No preamble, start your answer right away please
+# """
+
+# def process_inputs(audio_filepath, image_filepath):
+#     # Step 1: Speech to Text
+#     speech_to_text_output = transcribe_with_groq(
+#         GROQ_API_KEY=os.environ.get("GROQ_API_KEY"),
+#         audio_filepath=audio_filepath,
+#         stt_model="whisper-large-v3"
+#     )
+
+#     # Step 2: Vision + Reasoning
+#     if image_filepath:
+#         doctor_response = analyze_image_with_query(
+#             query=system_prompt + speech_to_text_output,
+#             encoded_image=encode_image(image_filepath),
+#             model="meta-llama/llama-4-scout-17b-16e-instruct"
+#         )
+#     else:
+#         doctor_response = "No image provided for me to analyze."
+
+#     # Step 3: Text to Speech (Doctor’s Voice)
+#     output_path = "final.mp3"
+#     text_to_speech_with_gtts(input_text=doctor_response, output_filepath=output_path)
+#     # Or you can switch to ElevenLabs if available:
+#     # text_to_speech_with_elevenlabs(input_text=doctor_response, output_filepath=output_path)
+
+#     # Step 4: Return outputs for Gradio
+#     return speech_to_text_output, doctor_response, output_path
+
+
+# # Step 5: Gradio Interface
+# iface = gr.Interface(
+#     fn=process_inputs,
+#     inputs=[
+#         gr.Audio(sources=["microphone"], type="filepath", label="Speak Your Symptoms"),
+#         gr.Image(type="filepath", label="Upload an Affected Area Image")
+#     ],
+#     outputs=[
+#         gr.Textbox(label="Speech to Text (What You Said)"),
+#         gr.Textbox(label="Doctor's Response"),
+#         gr.Audio(label="Doctor's Voice Output")
+#     ],
+#     title="🩺 AI Doctor with Vision and Voice",
+#     description="Speak your symptoms and upload an image — get a voice and text response from your AI Doctor.",
+#     theme="default"
+# )
+
+# if __name__ == "__main__":
+#     iface.launch(debug=True)
+
+
+# --------------------------------------------------------------------------------------------------------------------------
+import gradio as gr
+import os
+import pickle
+import numpy as np
+
+# -------------------
+# Load Models
+# -------------------
+working_dir = os.path.dirname(os.path.abspath(__file__))
+
+diabetes_model = pickle.load(open(f"{working_dir}/models/diabetes.pkl", "rb"))
+heart_model = pickle.load(open(f"{working_dir}/models/heart.pkl", "rb"))
+
+# TEMPORARY: Tumor model disabled for deployment
+# from tensorflow.keras.models import load_model
+# tumor_model = load_model(f"{working_dir}/models/model.h5")
+
+
+# -------------------
+# AI Doctor Function
+# -------------------
+from brain_of_the_doctor import encode_image, analyze_image_with_query
+from voice_of_the_patient import transcribe_with_groq
+from voice_of_the_doctor import text_to_speech_with_gtts
+
+system_prompt = """
+You have to act as a professional doctor, i know you are not but this is for learning purpose. 
+What's in this image? Do you find anything wrong with it medically? 
+If you make a differential, suggest some remedies for them. Donot add any numbers or special characters in your response. 
+Your response should be in one long paragraph. Also always answer as if you are answering to a real person.
+Donot say 'In the image I see' but say 'With what I see, I think you have ....'
+Dont respond as an AI model in markdown, your answer should mimic that of an actual doctor not an AI bot, 
+Keep your answer concise (max 2 sentences). No preamble, start your answer right away please
+"""
+
+def ai_doctor(audio_filepath, image_filepath):
+    speech_to_text_output = transcribe_with_groq(
+        GROQ_API_KEY=os.environ.get("GROQ_API_KEY"),
+        audio_filepath=audio_filepath,
+        stt_model="whisper-large-v3"
+    )
+    if image_filepath:
+        doctor_response = analyze_image_with_query(
+            query=system_prompt + speech_to_text_output,
+            encoded_image=encode_image(image_filepath),
+            model="meta-llama/llama-4-scout-17b-16e-instruct"
+        )
+    else:
+        doctor_response = "No image provided for analysis."
+
+    output_path = "final.mp3"
+    text_to_speech_with_gtts(input_text=doctor_response, output_filepath=output_path)
+    return speech_to_text_output, doctor_response, output_path
+
+
+# -------------------
+# ML Prediction Functions
+# -------------------
+def diabetes_predict(Pregnancies, Glucose, BloodPressure, SkinThickness, Insulin, BMI, DPF, Age):
+    user_input = [float(x) for x in [Pregnancies, Glucose, BloodPressure, SkinThickness, Insulin, BMI, DPF, Age]]
+    pred = diabetes_model.predict([user_input])[0]
+    return "Diabetic" if pred == 1 else "Not Diabetic"
+
+
+def heart_predict(age, sex, cp, trestbps, chol, fbs, restecg, thalach, exang, oldpeak, slope, ca, thal):
+    user_input = [float(x) for x in [age, sex, cp, trestbps, chol, fbs, restecg, thalach, exang, oldpeak, slope, ca, thal]]
+    pred = heart_model.predict([user_input])[0]
+    return "Heart Disease" if pred == 1 else "No Heart Disease"
+
+
+# TEMP Dummy Tumor prediction (TensorFlow removed)
+def tumor_predict(image):
+    return "Tumor model disabled temporarily – ONNX version will be added soon."
+
+
+# -------------------
+# Gradio Blocks App UI
+# -------------------
+with gr.Blocks(css="""
+body {background-color: #e6f2ff; font-family: 'Arial', sans-serif;}
+.gr-button {background: linear-gradient(to right, #4CAF50, #45a049) !important; color: white !important; font-weight: bold;}
+.gr-textbox {background-color: #ffffff !important; border-radius: 10px; padding: 8px; box-shadow: 0px 2px 5px rgba(0,0,0,0.1);}
+.gr-label {font-weight: bold; font-size: 14px; color: #333;}
+.gr-tabs-header {font-weight: bold; font-size: 16px; color: #333;}
+""") as demo:
+
+    with gr.Tabs():
+
+        # ------------------- AI Doctor -------------------
+        with gr.TabItem("AI Doctor"):
+            with gr.Row():
+                with gr.Column(scale=1):
+                    gr.Markdown("### Speak your symptoms & upload image")
+                    audio_input = gr.Audio(sources=["microphone"], type="filepath")
+                    image_input = gr.Image(type="filepath")
+                    doctor_button = gr.Button("Get Doctor Response")
+                with gr.Column(scale=1):
+                    st_text_output = gr.Textbox(label="Speech to Text", interactive=False)
+                    doctor_text_output = gr.Textbox(label="Doctor's Response", interactive=False)
+                    doctor_voice_output = gr.Audio(label="Doctor Voice", interactive=False)
+            doctor_button.click(ai_doctor, inputs=[audio_input, image_input],
+                                outputs=[st_text_output, doctor_text_output, doctor_voice_output])
+
+        # ------------------- Diabetes -------------------
+        with gr.TabItem("Diabetes Prediction"):
+            with gr.Row():
+                with gr.Column():
+                    Pregnancies = gr.Textbox(label="Pregnancies")
+                    Glucose = gr.Textbox(label="Glucose")
+                    BloodPressure = gr.Textbox(label="Blood Pressure")
+                    SkinThickness = gr.Textbox(label="Skin Thickness")
+                    Insulin = gr.Textbox(label="Insulin")
+                    BMI = gr.Textbox(label="BMI")
+                    DPF = gr.Textbox(label="Diabetes Pedigree Function")
+                    Age = gr.Textbox(label="Age")
+                    diabetes_button = gr.Button("Check Diabetes")
+                with gr.Column():
+                    diabetes_output = gr.Textbox(label="Result", interactive=False)
+            diabetes_button.click(diabetes_predict,
+                                  inputs=[Pregnancies, Glucose, BloodPressure, SkinThickness, Insulin, BMI, DPF, Age],
+                                  outputs=diabetes_output)
+
+        # ------------------- Heart -------------------
+        with gr.TabItem("Heart Prediction"):
+            with gr.Row():
+                with gr.Column():
+                    age = gr.Textbox(label="Age")
+                    sex = gr.Textbox(label="Sex")
+                    cp = gr.Textbox(label="Chest Pain Types")
+                    trestbps = gr.Textbox(label="Resting BP")
+                    chol = gr.Textbox(label="Cholesterol")
+                    fbs = gr.Textbox(label="Fasting Blood Sugar")
+                    restecg = gr.Textbox(label="Resting ECG")
+                    thalach = gr.Textbox(label="Max Heart Rate")
+                    exang = gr.Textbox(label="Exercise Induced Angina")
+                    oldpeak = gr.Textbox(label="ST Depression")
+                    slope = gr.Textbox(label="Slope of ST Segment")
+                    ca = gr.Textbox(label="Major Vessels")
+                    thal = gr.Textbox(label="Thalassemia")
+                    heart_button = gr.Button("Check Heart Disease")
+                with gr.Column():
+                    heart_output = gr.Textbox(label="Result", interactive=False)
+            heart_button.click(heart_predict,
+                               inputs=[age, sex, cp, trestbps, chol, fbs, restecg, thalach, exang, oldpeak, slope, ca, thal],
+                               outputs=heart_output)
+
+        # ------------------- Tumor -------------------
+        with gr.TabItem("Tumor Prediction"):
+            with gr.Row():
+                with gr.Column():
+                    tumor_image = gr.Image(type="filepath")
+                    tumor_button = gr.Button("Check Tumor")
+                with gr.Column():
+                    tumor_output = gr.Textbox(label="Result", interactive=False)
+            tumor_button.click(tumor_predict, inputs=tumor_image, outputs=tumor_output)
+
+if __name__ == "__main__":
+    demo.launch()