Upload 5 files

added project files of HEALIO

app.py

@@ -0,0 +1,561 @@
+from dotenv import load_dotenv
+load_dotenv()
+ 
+import os
+import gradio as gr
+ 
+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, text_to_speech_with_elevenlabs
+ 
+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 = ""
+    if audio_filepath:
+        try:
+            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"
+            )
+        except Exception as e:
+            speech_to_text_output = f"Error transcribing audio: {str(e)}"
+ 
+    if image_filepath:
+        try:
+            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"
+            )
+        except Exception as e:
+            doctor_response = f"Error analyzing image: {str(e)}"
+    else:
+        doctor_response = "No image provided for me to analyze."
+ 
+    output_filepath = "final.mp3"
+    try:
+        elevenlabs_key = os.environ.get("ELEVENLABS_API_KEY")
+        if elevenlabs_key:
+            text_to_speech_with_elevenlabs(input_text=doctor_response, output_filepath=output_filepath)
+        else:
+            text_to_speech_with_gtts(input_text=doctor_response, output_filepath=output_filepath)
+    except Exception as e:
+        print(f"ElevenLabs TTS failed, falling back to gTTS: {e}")
+        try:
+            text_to_speech_with_gtts(input_text=doctor_response, output_filepath=output_filepath)
+        except Exception as e2:
+            print(f"gTTS also failed: {e2}")
+            output_filepath = None
+ 
+    return speech_to_text_output, doctor_response, output_filepath
+ 
+ 
+
+# white/grey backgrounds from every Gradio wrapper div.
+# ─────────────────────────────────────────────────────────────────────────────
+ANIMATED_BG = """
+<style>
+  /* ── Step 1: force dark navy on root elements ── */
+  html, body {
+    background-color: #07111e !important;
+    background: #07111e !important;
+    margin: 0; padding: 0;
+  }
+ 
+  /* ── Step 2: make every Gradio shell transparent ── */
+  gradio-app,
+  gradio-app > div,
+  .gradio-container,
+  .gradio-container > div,
+  .main, .contain, .app,
+  #component-0, .tabs, .tabitem,
+  .form, .wrap, .gap, .svelte-1gfkn6j {
+    background: transparent !important;
+    background-color: transparent !important;
+  }
+ 
+  /* ── Step 3: pin canvas behind everything ── */
+  #med-canvas {
+    position: fixed;
+    top: 0; left: 0;
+    width: 100vw; height: 100vh;
+    z-index: 0;
+    pointer-events: none;
+  }
+ 
+  /* ── Step 4: lift Gradio content above canvas ── */
+  gradio-app {
+    position: relative;
+    z-index: 1;
+  }
+</style>
+ 
+<canvas id="med-canvas"></canvas>
+ 
+<script>
+(function () {
+  const canvas = document.getElementById('med-canvas');
+  const ctx    = canvas.getContext('2d');
+ 
+  function resize() {
+    canvas.width  = window.innerWidth;
+    canvas.height = window.innerHeight;
+  }
+  resize();
+  window.addEventListener('resize', resize);
+ 
+  /* ---------- Background gradient ---------- */
+  function drawBg() {
+    const g = ctx.createLinearGradient(0, 0, canvas.width, canvas.height);
+    g.addColorStop(0,    '#07111e');
+    g.addColorStop(0.5,  '#0c2040');
+    g.addColorStop(1,    '#061520');
+    ctx.fillStyle = g;
+    ctx.fillRect(0, 0, canvas.width, canvas.height);
+  }
+ 
+  /* ---------- Neural network ---------- */
+  const nodes = Array.from({ length: 44 }, () => ({
+    x:  Math.random() * window.innerWidth,
+    y:  Math.random() * window.innerHeight,
+    vx: (Math.random() - 0.5) * 0.28,
+    vy: (Math.random() - 0.5) * 0.28,
+    r:  1.6 + Math.random() * 2.4,
+    t:  Math.random() * Math.PI * 2,
+  }));
+ 
+  function stepNodes() {
+    nodes.forEach(n => {
+      n.x += n.vx; n.y += n.vy; n.t += 0.015;
+      if (n.x < 0 || n.x > canvas.width)  n.vx *= -1;
+      if (n.y < 0 || n.y > canvas.height) n.vy *= -1;
+    });
+  }
+ 
+  function drawNetwork() {
+    for (let i = 0; i < nodes.length; i++) {
+      for (let j = i + 1; j < nodes.length; j++) {
+        const dx = nodes[i].x - nodes[j].x;
+        const dy = nodes[i].y - nodes[j].y;
+        const d  = Math.hypot(dx, dy);
+        if (d < 155) {
+          ctx.beginPath();
+          ctx.moveTo(nodes[i].x, nodes[i].y);
+          ctx.lineTo(nodes[j].x, nodes[j].y);
+          ctx.strokeStyle = `rgba(65,160,205,${0.10 * (1 - d / 155)})`;
+          ctx.lineWidth   = 0.7;
+          ctx.stroke();
+        }
+      }
+    }
+    nodes.forEach(n => {
+      const a = 0.38 + 0.24 * Math.sin(n.t);
+      ctx.beginPath();
+      ctx.arc(n.x, n.y, n.r, 0, Math.PI * 2);
+      ctx.fillStyle = `rgba(88,182,222,${a})`;
+      ctx.fill();
+    });
+  }
+ 
+  /* ---------- Floating medical crosses ---------- */
+  const crosses = Array.from({ length: 9 }, () => ({
+    x:     Math.random() * window.innerWidth,
+    y:     Math.random() * window.innerHeight,
+    size:  10 + Math.random() * 18,
+    vy:    0.14 + Math.random() * 0.18,
+    alpha: 0.07 + Math.random() * 0.10,
+    angle: Math.random() * Math.PI * 2,
+    da:    (Math.random() - 0.5) * 0.003,
+  }));
+ 
+  function drawCrosses() {
+    crosses.forEach(c => {
+      c.y -= c.vy; c.angle += c.da;
+      if (c.y + c.size < 0) {
+        c.y = canvas.height + c.size;
+        c.x = Math.random() * canvas.width;
+      }
+      ctx.save();
+      ctx.translate(c.x, c.y);
+      ctx.rotate(c.angle);
+      ctx.strokeStyle = `rgba(100,188,226,${c.alpha})`;
+      ctx.lineWidth   = c.size * 0.20;
+      ctx.lineCap     = 'round';
+      ctx.beginPath(); ctx.moveTo(-c.size/2, 0); ctx.lineTo(c.size/2, 0); ctx.stroke();
+      ctx.beginPath(); ctx.moveTo(0, -c.size/2); ctx.lineTo(0, c.size/2); ctx.stroke();
+      ctx.restore();
+    });
+  }
+ 
+  /* ---------- DNA helices ---------- */
+  let dnaT = 0;
+  function drawDNA(cx) {
+    const amp = 26, steps = 22;
+    for (let i = 0; i < steps; i++) {
+      const y  = (i / steps) * canvas.height;
+      const ph = dnaT + i * 0.52;
+      const x1 = cx + amp * Math.sin(ph);
+      const x2 = cx - amp * Math.sin(ph);
+      ctx.beginPath(); ctx.arc(x1, y, 2.1, 0, Math.PI*2);
+      ctx.fillStyle = 'rgba(82,178,218,0.26)'; ctx.fill();
+      ctx.beginPath(); ctx.arc(x2, y, 2.1, 0, Math.PI*2);
+      ctx.fillStyle = 'rgba(82,178,218,0.26)'; ctx.fill();
+      if (i % 2 === 0) {
+        ctx.beginPath(); ctx.moveTo(x1, y); ctx.lineTo(x2, y);
+        ctx.strokeStyle = 'rgba(82,178,218,0.11)'; ctx.lineWidth = 1; ctx.stroke();
+      }
+    }
+    dnaT += 0.009;
+  }
+ 
+  /* ---------- ECG heartbeat line ---------- */
+  let ecgOff = 0;
+  function ecgY(x) {
+    const t    = (x + ecgOff) * 0.040;
+    const base = Math.sin(t * 0.60) * 5;
+    const mod  = x % 128;
+    const spike = mod < 8 ? (mod < 4 ? mod * 7 : (8 - mod) * 7) - 4 : 0;
+    return base + spike * 2.8;
+  }
+  function drawECG() {
+    ecgOff += 1.0;
+    const y0 = canvas.height * 0.91;
+    ctx.beginPath();
+    for (let x = 0; x <= canvas.width; x++) {
+      const y = y0 + ecgY(x);
+      x === 0 ? ctx.moveTo(x, y) : ctx.lineTo(x, y);
+    }
+    ctx.strokeStyle = 'rgba(100,196,232,0.26)';
+    ctx.lineWidth   = 1.4;
+    ctx.stroke();
+  }
+ 
+  /* ---------- Expanding pulse rings ---------- */
+  const rings = [0, 0.33, 0.66].map(p => ({ p }));
+  function drawRings() {
+    const cx = canvas.width * 0.5, cy = canvas.height * 0.44;
+    rings.forEach(r => {
+      r.p = (r.p + 0.0033) % 1;
+      ctx.beginPath();
+      ctx.arc(cx, cy, 70 + 265 * r.p, 0, Math.PI * 2);
+      ctx.strokeStyle = `rgba(92,188,226,${0.046 * (1 - r.p)})`;
+      ctx.lineWidth   = 1.1;
+      ctx.stroke();
+    });
+  }
+ 
+  /* ---------- Main loop ---------- */
+  function loop() {
+    drawBg();
+    drawRings();
+    drawDNA(46);
+    drawDNA(canvas.width - 46);
+    stepNodes();
+    drawNetwork();
+    drawCrosses();
+    drawECG();
+    requestAnimationFrame(loop);
+  }
+ 
+  // Small delay lets Gradio finish mounting before we start
+  setTimeout(loop, 150);
+})();
+</script>
+"""
+ 
+# ── CSS ───────────────────────────────────────────────────────────────────────
+custom_css = """
+@import url('https://fonts.googleapis.com/css2?family=Orbitron:wght@400;600;700&family=Exo+2:wght@300;400;500;600&display=swap');
+ 
+:root {
+  --panel-bg:    rgba(8, 26, 52, 0.74);
+  --border-soft: rgba(90, 180, 220, 0.22);
+  --text-main:   #ddf0fa;
+  --text-muted:  #7ab8d4;
+}
+ 
+/* Ensure dark base at CSS level too */
+html { background: #07111e !important; }
+body { background: #07111e !important; }
+ 
+gradio-app, gradio-app > div,
+.gradio-container, .gradio-container > div,
+.main, .contain, .app, #component-0,
+.tabs, .tabitem, .form, .wrap, .gap {
+  background: transparent !important;
+  background-color: transparent !important;
+}
+ 
+body, gradio-app {
+  font-family: 'Exo 2', sans-serif !important;
+  color: var(--text-main) !important;
+}
+ 
+/* Title */
+h1 {
+  font-family: 'Orbitron', monospace !important;
+  font-weight: 700 !important;
+  font-size: clamp(1.3rem, 2.6vw, 2rem) !important;
+  letter-spacing: 0.10em !important;
+  text-align: center !important;
+  color: #8dd8f0 !important;
+  padding: 1.4rem 0 0.2rem !important;
+}
+h1::after {
+  content: '';
+  display: block;
+  margin: 0.45rem auto 0;
+  width: 200px; height: 1.5px;
+  background: linear-gradient(90deg, transparent, #5ab4dc, #4ecdc4, transparent);
+  border-radius: 2px;
+}
+ 
+/* Cards */
+.block, .gr-block, .gr-box, .gr-panel, .panel {
+  background: var(--panel-bg) !important;
+  border: 1px solid var(--border-soft) !important;
+  border-radius: 12px !important;
+  backdrop-filter: blur(20px) saturate(1.4) !important;
+  -webkit-backdrop-filter: blur(20px) saturate(1.4) !important;
+  box-shadow: 0 4px 28px rgba(0,0,0,0.55) !important;
+  transition: border-color 0.3s !important;
+}
+.block:hover { border-color: rgba(90,180,220,0.42) !important; }
+ 
+/* Labels */
+label span, .gr-label, label {
+  font-family: 'Exo 2', sans-serif !important;
+  font-size: 0.76rem !important;
+  font-weight: 600 !important;
+  letter-spacing: 0.08em !important;
+  color: #7ecfe8 !important;
+  text-transform: uppercase !important;
+}
+ 
+/* Inputs */
+textarea, input[type="text"] {
+  background: rgba(4,16,36,0.82) !important;
+  border: 1px solid rgba(90,180,220,0.20) !important;
+  border-radius: 8px !important;
+  color: var(--text-main) !important;
+  font-family: 'Exo 2', sans-serif !important;
+  font-size: 0.90rem !important;
+  transition: border-color 0.2s !important;
+}
+textarea:focus, input[type="text"]:focus {
+  border-color: rgba(90,180,220,0.50) !important;
+  outline: none !important;
+}
+ 
+/* Primary button */
+button.primary, .gr-button-primary, button[variant="primary"] {
+  font-family: 'Orbitron', monospace !important;
+  font-weight: 600 !important;
+  font-size: 0.73rem !important;
+  letter-spacing: 0.09em !important;
+  text-transform: uppercase !important;
+  background: linear-gradient(135deg, #0d4a6e 0%, #1a7aab 100%) !important;
+  color: #ddf0fa !important;
+  border: 1px solid rgba(90,180,220,0.36) !important;
+  border-radius: 8px !important;
+  padding: 0.6rem 1.5rem !important;
+  transition: background 0.25s, transform 0.15s !important;
+}
+button.primary:hover {
+  background: linear-gradient(135deg, #115880, #2192cc) !important;
+  transform: translateY(-1px) !important;
+}
+button.secondary, .gr-button-secondary {
+  font-family: 'Exo 2', sans-serif !important;
+  font-size: 0.77rem !important;
+  background: rgba(90,180,220,0.07) !important;
+  color: #7ecfe8 !important;
+  border: 1px solid rgba(90,180,220,0.22) !important;
+  border-radius: 8px !important;
+  transition: background 0.2s !important;
+}
+button.secondary:hover { background: rgba(90,180,220,0.16) !important; }
+ 
+/* Audio */
+.waveform-container, audio {
+  background: rgba(4,16,36,0.65) !important;
+  border-radius: 10px !important;
+  border: 1px solid rgba(90,180,220,0.16) !important;
+}
+ 
+/* Image upload */
+.image-container, .upload-container {
+  border: 1.5px dashed rgba(90,180,220,0.25) !important;
+  border-radius: 10px !important;
+  background: rgba(4,16,36,0.50) !important;
+  transition: border-color 0.3s !important;
+}
+.image-container:hover { border-color: rgba(90,180,220,0.50) !important; }
+ 
+::-webkit-scrollbar { width: 5px; }
+::-webkit-scrollbar-track { background: rgba(0,16,32,0.4); }
+::-webkit-scrollbar-thumb { background: #2a6e96; border-radius: 3px; }
+ 
+footer { display: none !important; }
+"""
+ 
+# ── Gradio theme ──────────────────────────────────────────────────────────────
+theme = gr.themes.Base(
+    primary_hue=gr.themes.colors.cyan,
+    secondary_hue=gr.themes.colors.blue,
+    neutral_hue=gr.themes.colors.slate,
+    font=[gr.themes.GoogleFont("Exo 2"), "sans-serif"],
+    font_mono=[gr.themes.GoogleFont("Orbitron"), "monospace"],
+).set(
+    body_background_fill="transparent",
+    body_background_fill_dark="transparent",
+    block_background_fill="rgba(8,26,52,0.74)",
+    block_background_fill_dark="rgba(8,26,52,0.74)",
+    block_border_color="rgba(90,180,220,0.22)",
+    block_border_color_dark="rgba(90,180,220,0.22)",
+    block_label_text_color="#7ecfe8",
+    block_label_text_color_dark="#7ecfe8",
+    input_background_fill="rgba(4,16,36,0.82)",
+    input_background_fill_dark="rgba(4,16,36,0.82)",
+    input_border_color="rgba(90,180,220,0.20)",
+    input_border_color_dark="rgba(90,180,220,0.20)",
+    button_primary_background_fill="linear-gradient(135deg,#0d4a6e,#1a7aab)",
+    button_primary_background_fill_dark="linear-gradient(135deg,#0d4a6e,#1a7aab)",
+    button_primary_text_color="#ddf0fa",
+    button_primary_text_color_dark="#ddf0fa",
+)
+ 
+# ── UI ────────────────────────────────────────────────────────────────────────
+with gr.Blocks(theme=theme, css=custom_css, title="AI Doctor — Vision & Voice") as iface:
+ 
+    gr.HTML(ANIMATED_BG)
+ 
+    gr.Markdown("""
+<h1>⚕ Healio-AI Doctor Vision &amp; Voice</h1>
+<p style='text-align:center;color:#7ab8d4;font-family:Exo 2,sans-serif;
+font-size:0.91rem;letter-spacing:0.03em;margin-bottom:0.6rem;'>
+Upload a medical image and describe your symptoms via microphone.<br>
+The AI doctor will analyze and respond with a diagnosis.
+</p>
+""")
+ 
+    # Cute animated AI doctor robot in the bottom-right corner
+    gr.HTML("""
+    <style>
+      #ai-doc-robot {
+        position: fixed;
+        bottom: 18px;
+        right: 22px;
+        z-index: 999;
+        width: 80px;
+        cursor: pointer;
+        filter: drop-shadow(0 4px 12px rgba(0,180,220,0.35));
+        animation: robot-float 3.2s ease-in-out infinite;
+      }
+      @keyframes robot-float {
+        0%,100% { transform: translateY(0px);   }
+        50%      { transform: translateY(-10px); }
+      }
+      #ai-doc-robot:hover { animation: robot-wiggle 0.5s ease-in-out; }
+      @keyframes robot-wiggle {
+        0%,100% { transform: rotate(0deg);   }
+        25%     { transform: rotate(-8deg);  }
+        75%     { transform: rotate(8deg);   }
+      }
+    </style>
+    <svg id="ai-doc-robot" viewBox="0 0 80 100" xmlns="http://www.w3.org/2000/svg">
+      <!-- Antenna -->
+      <line x1="40" y1="8" x2="40" y2="18" stroke="#5ab4dc" stroke-width="2.5" stroke-linecap="round"/>
+      <circle cx="40" cy="6" r="4" fill="#4ecdc4" opacity="0.9">
+        <animate attributeName="opacity" values="0.9;0.4;0.9" dur="1.4s" repeatCount="indefinite"/>
+        <animate attributeName="r" values="4;5.5;4" dur="1.4s" repeatCount="indefinite"/>
+      </circle>
+      <!-- Head -->
+      <rect x="20" y="18" width="40" height="32" rx="9" fill="#0d3a58" stroke="#5ab4dc" stroke-width="1.5"/>
+      <!-- Eyes -->
+      <ellipse cx="31" cy="31" rx="6" ry="6" fill="#07111e"/>
+      <ellipse cx="49" cy="31" rx="6" ry="6" fill="#07111e"/>
+      <circle cx="31" cy="31" r="3.5" fill="#4ecdc4">
+        <animate attributeName="r" values="3.5;2;3.5" dur="3s" repeatCount="indefinite"/>
+      </circle>
+      <circle cx="49" cy="31" r="3.5" fill="#4ecdc4">
+        <animate attributeName="r" values="3.5;2;3.5" dur="3s" begin="0.3s" repeatCount="indefinite"/>
+      </circle>
+      <!-- Eye shine -->
+      <circle cx="33" cy="29" r="1.2" fill="white" opacity="0.7"/>
+      <circle cx="51" cy="29" r="1.2" fill="white" opacity="0.7"/>
+      <!-- Mouth / display bar -->
+      <rect x="27" y="41" width="26" height="5" rx="2.5" fill="#07111e" stroke="#5ab4dc" stroke-width="0.8"/>
+      <rect x="29" y="42.5" width="6" height="2" rx="1" fill="#4ecdc4">
+        <animate attributeName="width" values="6;14;6" dur="1.2s" repeatCount="indefinite"/>
+      </rect>
+      <!-- Neck -->
+      <rect x="36" y="50" width="8" height="6" rx="2" fill="#0a2e48"/>
+      <!-- Body -->
+      <rect x="16" y="56" width="48" height="34" rx="10" fill="#0d3a58" stroke="#5ab4dc" stroke-width="1.5"/>
+      <!-- Chest cross / medical -->
+      <rect x="37" y="63" width="6" height="14" rx="2" fill="#4ecdc4" opacity="0.85"/>
+      <rect x="33" y="67" width="14" height="6" rx="2" fill="#4ecdc4" opacity="0.85"/>
+      <!-- Side panel dots -->
+      <circle cx="23" cy="67" r="2.5" fill="#5ab4dc" opacity="0.6">
+        <animate attributeName="opacity" values="0.6;1;0.6" dur="1.8s" repeatCount="indefinite"/>
+      </circle>
+      <circle cx="23" cy="75" r="2.5" fill="#5ab4dc" opacity="0.6">
+        <animate attributeName="opacity" values="0.6;1;0.6" dur="1.8s" begin="0.6s" repeatCount="indefinite"/>
+      </circle>
+      <circle cx="57" cy="67" r="2.5" fill="#5ab4dc" opacity="0.6">
+        <animate attributeName="opacity" values="0.6;1;0.6" dur="1.8s" begin="0.3s" repeatCount="indefinite"/>
+      </circle>
+      <circle cx="57" cy="75" r="2.5" fill="#5ab4dc" opacity="0.6">
+        <animate attributeName="opacity" values="0.6;1;0.6" dur="1.8s" begin="0.9s" repeatCount="indefinite"/>
+      </circle>
+      <!-- Arms -->
+      <rect x="2" y="58" width="14" height="8" rx="4" fill="#0d3a58" stroke="#5ab4dc" stroke-width="1.2">
+        <animateTransform attributeName="transform" type="rotate" values="0 9 62; 15 9 62; 0 9 62" dur="3.2s" repeatCount="indefinite"/>
+      </rect>
+      <rect x="64" y="58" width="14" height="8" rx="4" fill="#0d3a58" stroke="#5ab4dc" stroke-width="1.2">
+        <animateTransform attributeName="transform" type="rotate" values="0 71 62; -15 71 62; 0 71 62" dur="3.2s" repeatCount="indefinite"/>
+      </rect>
+      <!-- Legs -->
+      <rect x="25" y="90" width="12" height="8" rx="4" fill="#0a2e48" stroke="#5ab4dc" stroke-width="1"/>
+      <rect x="43" y="90" width="12" height="8" rx="4" fill="#0a2e48" stroke="#5ab4dc" stroke-width="1"/>
+    </svg>
+    """)
+ 
+    with gr.Row():
+        with gr.Column(scale=1):
+            audio_input = gr.Audio(sources=["microphone"], type="filepath",
+                                   label="🎙  Patient Audio Input")
+            image_input = gr.Image(type="filepath",
+                                   label="🩻  Medical Image Upload")
+            with gr.Row():
+                submit_btn = gr.Button("🩺  Analyze", variant="primary")
+                clear_btn  = gr.Button("🗑  Clear",   variant="secondary")
+ 
+        with gr.Column(scale=1):
+            text_out  = gr.Textbox(label="📝  Speech Transcription", lines=2)
+            doc_out   = gr.Textbox(label="🩺  Doctor's Diagnosis",   lines=5)
+            audio_out = gr.Audio(label="🔊  Doctor's Voice Response", type="filepath")
+ 
+    submit_btn.click(
+        fn=process_inputs,
+        inputs=[audio_input, image_input],
+        outputs=[text_out, doc_out, audio_out],
+    )
+ 
+    clear_btn.click(
+        fn=lambda: (None, None, "", "", None),
+        inputs=[],
+        outputs=[audio_input, image_input, text_out, doc_out, audio_out],
+    )
+ 
+iface.launch(server_name="0.0.0.0",server_port=7860, allowed_paths=["."])

brain_of_the_doctor.py

@@ -0,0 +1,39 @@
+from dotenv import load_dotenv
+load_dotenv()
+ 
+import os
+import base64
+from groq import Groq
+ 
+GROQ_API_KEY = os.environ.get("GROQ_API_KEY")
+ 
+ 
+def encode_image(image_path):
+    with open(image_path, "rb") as image_file:
+        return base64.b64encode(image_file.read()).decode('utf-8')
+ 
+ 
+def analyze_image_with_query(query, model, encoded_image):
+    client = Groq(api_key=GROQ_API_KEY)
+    messages = [
+        {
+            "role": "user",
+            "content": [
+                {
+                    "type": "text",
+                    "text": query
+                },
+                {
+                    "type": "image_url",
+                    "image_url": {
+                        "url": f"data:image/jpeg;base64,{encoded_image}",
+                    },
+                },
+            ],
+        }
+    ]
+    chat_completion = client.chat.completions.create(
+        messages=messages,
+        model=model
+    )
+    return chat_completion.choices[0].message.content

requirements.txt

@@ -0,0 +1,10 @@
+gradio==5.12.0
+groq==0.15.0
+elevenlabs==1.50.3
+gtts==2.5.4
+speechrecognition==3.13.0
+pydub==0.25.1
+pillow==11.1.0
+requests==2.32.3
+python-dotenv
+httpx==0.28.1

voice_of_the_doctor.py

@@ -0,0 +1,75 @@
+from dotenv import load_dotenv
+load_dotenv()
+ 
+import os
+import subprocess
+import platform
+from gtts import gTTS
+from elevenlabs.client import ElevenLabs
+from elevenlabs import save
+ 
+ELEVENLABS_API_KEY = os.environ.get("ELEVENLABS_API_KEY")
+ 
+ 
+def text_to_speech_with_gtts(input_text, output_filepath):
+    """Convert text to speech using gTTS and save to file."""
+    language = "en"
+    audioobj = gTTS(
+        text=input_text,
+        lang=language,
+        slow=False
+    )
+    audioobj.save(output_filepath)
+ 
+    os_name = platform.system()
+    try:
+        if os_name == "Darwin":        # macOS
+            subprocess.run(['afplay', output_filepath])
+        elif os_name == "Windows":     # Windows
+            subprocess.run(
+                ['ffplay', '-nodisp', '-autoexit', output_filepath],
+                stdout=subprocess.DEVNULL,
+                stderr=subprocess.DEVNULL
+            )
+        elif os_name == "Linux":       # Linux
+            subprocess.run(['aplay', output_filepath])
+        else:
+            raise OSError("Unsupported operating system")
+    except Exception as e:
+        print(f"An error occurred while trying to play the audio: {e}")
+ 
+ 
+def text_to_speech_with_elevenlabs(input_text, output_filepath):
+    """Convert text to speech using ElevenLabs and save to file."""
+    client = ElevenLabs(api_key=ELEVENLABS_API_KEY)
+ 
+    # generate() returns a generator; collect all chunks before saving
+    audio_generator = client.text_to_speech.convert(
+        text=input_text,
+        voice_id="EXAVITQu4vr4xnSDxMaL",   # "Aria" voice ID
+        output_format="mp3_22050_32",
+        model_id="eleven_turbo_v2"
+    )
+ 
+    # Consume the generator and write bytes to file
+    with open(output_filepath, "wb") as f:
+        for chunk in audio_generator:
+            if chunk:
+                f.write(chunk)
+ 
+    os_name = platform.system()
+    try:
+        if os_name == "Darwin":        # macOS
+            subprocess.run(['afplay', output_filepath])
+        elif os_name == "Windows":     # Windows
+            subprocess.run(
+                ['ffplay', '-nodisp', '-autoexit', output_filepath],
+                stdout=subprocess.DEVNULL,
+                stderr=subprocess.DEVNULL
+            )
+        elif os_name == "Linux":       # Linux
+            subprocess.run(['aplay', output_filepath])
+        else:
+            raise OSError("Unsupported operating system")
+    except Exception as e:
+        print(f"An error occurred while trying to play the audio: {e}")

voice_of_the_patient.py

@@ -0,0 +1,49 @@
+from dotenv import load_dotenv
+load_dotenv()
+ 
+import os
+import logging
+import speech_recognition as sr
+from pydub import AudioSegment
+from io import BytesIO
+from groq import Groq
+ 
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
+ 
+GROQ_API_KEY = os.environ.get("GROQ_API_KEY")
+ 
+ 
+def record_audio(file_path, timeout=20, phrase_time_limit=None):
+    """Record audio from microphone and save as MP3."""
+    recognizer = sr.Recognizer()
+ 
+    try:
+        with sr.Microphone() as source:
+            logging.info("Adjusting for ambient noise...")
+            recognizer.adjust_for_ambient_noise(source, duration=1)
+            logging.info("Start speaking now...")
+ 
+            audio_data = recognizer.listen(source, timeout=timeout, phrase_time_limit=phrase_time_limit)
+            logging.info("Recording complete.")
+ 
+            wav_data = audio_data.get_wav_data()
+            audio_segment = AudioSegment.from_wav(BytesIO(wav_data))
+            audio_segment.export(file_path, format="mp3", bitrate="128k")
+ 
+            logging.info(f"Audio saved to {file_path}")
+ 
+    except Exception as e:
+        logging.error(f"An error occurred: {e}")
+ 
+ 
+def transcribe_with_groq(stt_model, audio_filepath, GROQ_API_KEY):
+    """Transcribe audio file using Groq Whisper."""
+    client = Groq(api_key=GROQ_API_KEY)
+ 
+    with open(audio_filepath, "rb") as audio_file:
+        transcription = client.audio.transcriptions.create(
+            model=stt_model,
+            file=audio_file,
+            language="en"
+        )
+    return transcription.text