Create app.py

app.py

@@ -0,0 +1,54 @@
+import os
+import whisper
+import gradio as gr
+import requests
+from gtts import gTTS
+
+# Load Whisper model
+model = whisper.load_model("base")
+
+# Read Groq API Key from environment variable
+GROQ_API_KEY = os.getenv("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")
+client= GROQ(API_KEY=GROQ_API_KEY)
+# Main function: audio → text → LLM → speech
+def transcribe_and_respond(audio_file):
+    # 1. Transcribe audio
+    result = model.transcribe(audio_file)
+    user_text = result["text"]
+
+    # 2. Query Groq LLM
+    headers = {
+        "Content-Type": "application/json",
+        "Authorization": f"Bearer {GROQ_API_KEY}"
+    }
+
+    data = {
+        "model": "llama-3.3-70b-versatile",
+        "messages": [{"role": "user", "content": user_text}]
+    }
+
+    response = requests.post("https://api.groq.com/openai/v1/chat/completions", headers=headers, json=data)
+
+    if response.status_code == 200:
+        output_text = response.json()['choices'][0]['message']['content']
+    else:
+        output_text = f"Error from Groq API: {response.status_code} - {response.text}"
+
+    # 3. Convert to speech
+    tts = gTTS(text=output_text, lang='en')
+    tts_path = "response.mp3"
+    tts.save(tts_path)
+
+    return output_text, tts_path
+
+# Gradio UI
+iface = gr.Interface(
+    fn=transcribe_and_respond,
+    inputs=gr.Audio(type="filepath", label="🎙️ Speak"),
+    outputs=[gr.Textbox(label="🧠 LLM Reply"), gr.Audio(label="🔊 Spoken Response")],
+    title="Voice Chatbot with Whisper + Groq + gTTS",
+    description="Click to record → Get LLM reply → Hear it spoken back"
+)
+
+if __name__ == "__main__":
+    iface.launch()