Create app.py

app.py

@@ -0,0 +1,138 @@
+import gradio as gr
+import random
+from google import generativeai as genai
+
+# Configure Gemini
+genai.configure(api_key="YOUR_API_KEY_HERE")
+model = genai.GenerativeModel("gemini-2.0-flash")
+
+# Evaluation logic
+def getscore(question, answer):
+    info = f"""
+You are an interview expert but be a little free not toooo strict. A user was asked the following interview question:
+
+Question: {question}
+
+Their answer was: "{answer}"
+
+Please do the following:
+1. Rate the answer from 1 to 10 based on relevance, clarity, and depth.
+2. Provide detailed feedback on how the answer can be improved in a single line.
+Respond in the format:
+1.  **Rating:** <score>/10
+2.  **Feedback:** <your feedback here>
+"""
+    response = model.generate_content(info)
+    text = response.text.strip()
+    lines = text.split("\n")
+    score = -1
+    feedback = "Feedback not found"
+    for line in lines:
+        if "**Rating:**" in line:
+            try:
+                score = int(line.split("**Rating:**")[1].split("/")[0].strip())
+            except:
+                score = -1
+        elif "**Feedback:**" in line:
+            feedback = line.split("**Feedback:**")[1].strip()
+    return [score, feedback]
+
+def finalsummary(feedback_list):
+    prompt = f"""
+You are given a list of feedback. Summarize it and give 5–6 final suggestions for improvement.
+feedback={feedback_list}
+"""
+    response = model.generate_content(prompt)
+    return response.text
+
+# Question bank (shortened example)
+imp_questions_map = {
+    "ml": [
+        "What is supervised learning? Give an example.",
+        "What is the difference between classification and regression?",
+        "How does linear regression work?",
+        "What are decision trees and how do they work?",
+        "What is overfitting vs underfitting?"
+    ],
+    "general": [
+        "Tell me about yourself.",
+        "What are your strengths and weaknesses?",
+        "How do you handle stress or pressure?",
+        "Describe a time when you worked in a team.",
+        "Why should we hire you?"
+    ]
+}
+
+# App state
+session = {
+    "questions": [],
+    "current_index": 0,
+    "answers": [],
+    "scores": [],
+    "feedbacks": []
+}
+
+# Main logic handler
+def start_session(field):
+    session["questions"] = random.sample(imp_questions_map.get(field, []), 5) + random.sample(imp_questions_map["general"], 3)
+    session["current_index"] = 0
+    session["answers"] = []
+    session["scores"] = []
+    session["feedbacks"] = []
+    return session["questions"][0]
+
+def submit_answer(audio):
+    import whisper
+    model = whisper.load_model("base")
+    audio_path = audio
+    result = model.transcribe(audio_path)
+    answer = result["text"]
+
+    # Evaluate
+    question = session["questions"][session["current_index"]]
+    score, fb = getscore(question, answer)
+    session["answers"].append(answer)
+    session["scores"].append(score)
+    session["feedbacks"].append(fb)
+
+    session["current_index"] += 1
+
+    if session["current_index"] < len(session["questions"]):
+        next_q = session["questions"][session["current_index"]]
+        return next_q, gr.update(visible=True), gr.update(visible=False), ""
+    else:
+        # Done - show summary
+        total_score = sum(session["scores"]) / len(session["scores"])
+        summary = finalsummary(session["feedbacks"])
+        return "Interview completed!", gr.update(visible=False), gr.update(visible=True), f"Average Score: {total_score:.2f}/10\n\nSummary:\n{summary}"
+
+def show_detailed_feedback():
+    detailed = ""
+    for i, q in enumerate(session["questions"]):
+        detailed += f"**Q{i+1}: {q}**\nAnswer: {session['answers'][i]}\nScore: {session['scores'][i]}/10\nFeedback: {session['feedbacks'][i]}\n\n"
+    return detailed
+
+# Gradio UI
+with gr.Blocks() as demo:
+    gr.Markdown("## AI Mock Interview Evaluator")
+
+    field_input = gr.Dropdown(
+        choices=list(imp_questions_map.keys()),
+        label="Select your field"
+    )
+
+    start_btn = gr.Button("Start Interview")
+    question_output = gr.Textbox(label="Question", interactive=False)
+    audio_input = gr.Audio(source="microphone", type="filepath", label="Record your answer")
+    submit_btn = gr.Button("Submit Answer")
+
+    final_output = gr.Textbox(label="Final Score and Summary", visible=False)
+    detailed_btn = gr.Button("Show Detailed Feedback", visible=False)
+    detailed_output = gr.Markdown()
+
+    start_btn.click(start_session, inputs=field_input, outputs=question_output)
+    submit_btn.click(submit_answer, inputs=audio_input,
+                     outputs=[question_output, audio_input, final_output, detailed_output])
+    detailed_btn.click(show_detailed_feedback, outputs=detailed_output)
+
+demo.launch()