Update app.py

app.py

@@ -0,0 +1,73 @@
+import os
+import gradio as gr
+import whisper
+from gtts import gTTS
+from groq import Groq
+import numpy as np
+
+
+
+# Set your Groq API key
+os.environ['GROQ_API_KEY'] = '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'
+
+# Initialize Whisper and Groq
+whisper_model = whisper.load_model("base")
+client = Groq(api_key=os.environ.get("GROQ_API_KEY"))
+
+def chatbot(audio_input):
+    try:
+        # Debug: Check the type and content of audio_input
+        print(f"Audio input type: {type(audio_input)}")
+        if audio_input is None:
+            raise ValueError("Audio input is None. Please provide a valid audio file.")
+
+        # Step 1: Transcribe audio input using OpenAI Whisper
+        transcription_result = whisper_model.transcribe(audio_input)
+        if transcription_result is None or not transcription_result.get("text"):
+            raise ValueError("Whisper transcription failed or returned empty text.")
+
+        transcription = transcription_result["text"]
+
+        # Step 2: Generate response using LLaMA 8B model via Groq API
+        chat_completion = client.chat.completions.create(
+            messages=[
+                {
+                    "role": "user",
+                    "content": transcription,
+                }
+            ],
+            model="llama3-8b-8192",
+        )
+
+        # Check if the response from Groq is valid
+        if chat_completion and chat_completion.choices:
+            response_text = chat_completion.choices[0].message.content
+        else:
+            raise ValueError("Invalid response from Groq API")
+
+        # Step 3: Convert text response to speech using GTTS
+        if response_text.strip():
+            tts = gTTS(response_text)
+            tts.save("response.mp3")
+        else:
+            raise ValueError("Response text is empty or invalid")
+
+        # Step 4: Return the response audio and text transcription
+        return "response.mp3", transcription, response_text
+
+    except Exception as e:
+        # Handle and display the error
+        return None, transcription if 'transcription' in locals() else None, f"Error: {str(e)}"
+
+# Define the Gradio interface
+interface = gr.Interface(
+    fn=chatbot,
+    inputs=gr.Audio(type="filepath"),
+    outputs=[gr.Audio(type="filepath"), "text", "text"],
+    title="Voice-to-Voice Chatbot",
+    description="Speak to the chatbot and get a real-time response.",
+    live=True  # Automatically processes input without requiring a button click
+)
+
+# Launch the Gradio app
+interface.launch()