Update app.py

app.py

@@ -1,87 +1,82 @@
 import gradio as gr
-import openai
+import whisper
 import os
+import uuid
 import pdfplumber
 import docx
 import pandas as pd
 from PIL import Image
-import pytesseract
-from pydub import AudioSegment
-import tempfile
-import sounddevice as sd
-import scipy.io.wavfile as wav
 
-openai.api_key = os.getenv("OPENAI_API_KEY")
+# Load Whisper model locally
+whisper_model = whisper.load_model("base")
 
-def extract_text(file):
-    ext = file.name.split(".")[-1].lower()
-    if ext == "pdf":
-        with pdfplumber.open(file.name) as pdf:
-            return "\n".join([page.extract_text() for page in pdf.pages if page.extract_text()])
-    elif ext in ["doc", "docx"]:
-        doc = docx.Document(file.name)
-        return "\n".join([p.text for p in doc.paragraphs])
-    elif ext in ["xls", "xlsx", "csv"]:
-        df = pd.read_excel(file.name) if ext != "csv" else pd.read_csv(file.name)
-        return df.to_string()
-    elif ext in ["jpg", "jpeg", "png"]:
-        image = Image.open(file.name)
-        text = pytesseract.image_to_string(image)
-        return text or "❌ No text found in image."
-    else:
-        return "❌ Unsupported file format."
+def transcribe_audio(audio):
+    if audio is None:
+        return "⚠️ No audio file uploaded."
 
-def transcribe_audio(audio_path):
     try:
-        with open(audio_path, "rb") as f:
-            transcript = openai.Audio.transcribe("whisper-1", f)
-        return transcript["text"]
+        result = whisper_model.transcribe(audio)
+        return result["text"]
     except Exception as e:
-        return f"❌ Transcription error: {str(e)}"
+        return f"❌ Whisper error: {str(e)}"
 
-def generate_response(messages):
-    try:
-        response = openai.ChatCompletion.create(
-            model="gpt-3.5-turbo",
-            messages=messages
-        )
-        return response.choices[0].message["content"]
-    except Exception as e:
-        return f"❌ Error: {str(e)}"
+def extract_text_from_file(file):
+    if file is None:
+        return "⚠️ No file uploaded."
 
-chat_history = []
+    name, ext = os.path.splitext(file.name)
+    ext = ext.lower()
 
-def chat(user_message, file=None, image=None, mic_audio=None):
-    if mic_audio:
-        user_message = transcribe_audio(mic_audio)
-    elif file:
-        user_message = extract_text(file)
-    elif image:
-        user_message = extract_text(image)
+    try:
+        if ext == ".pdf":
+            with pdfplumber.open(file) as pdf:
+                text = "\n".join(page.extract_text() or "" for page in pdf.pages)
+        elif ext in [".docx"]:
+            doc = docx.Document(file)
+            text = "\n".join(paragraph.text for paragraph in doc.paragraphs)
+        elif ext in [".xlsx", ".xls"]:
+            df = pd.read_excel(file)
+            text = df.to_string(index=False)
+        elif ext in [".png", ".jpg", ".jpeg"]:
+            image = Image.open(file)
+            text = "🖼️ Image uploaded. Please describe what you'd like me to do with it."
+        else:
+            text = "⚠️ Unsupported file type."
+        return text
+    except Exception as e:
+        return f"❌ File processing error: {str(e)}"
 
-    chat_history.append({"role": "user", "content": user_message})
-    bot_response = generate_response(chat_history)
-    chat_history.append({"role": "assistant", "content": bot_response})
-    return bot_response
+def chatbot_response(message, chat_history):
+    # Echo-style placeholder response
+    bot_reply = f"🤖 You said: {message}"
+    chat_history.append((message, bot_reply))
+    return "", chat_history
 
-with gr.Blocks() as demo:
-    gr.Markdown("### 🎧 **Neobot - Always Listening**")
-    chatbot = gr.Chatbot(height=300)
+with gr.Blocks(css="""
+#chatbox { height: 500px; overflow: auto; }
+.message-input { height: 40px; border-radius: 6px; }
+.sidebar { width: 25%; overflow-y: auto; }
+.main { width: 75%; }
+""") as demo:
+    gr.Markdown("""<h2 style="text-align:center;">🤖 Neobot - Always Listening</h2>""")
 
     with gr.Row():
-        txt = gr.Textbox(placeholder="Type here or use mic...", scale=4)
-        send_btn = gr.Button("🚀", scale=1)
+        with gr.Column(scale=1, min_width=200):
+            saved_chats = gr.Textbox(label="Saved Chats (auto-listed soon)", interactive=False)
+        with gr.Column(scale=4):
+            chatbot = gr.Chatbot(elem_id="chatbox")
 
     with gr.Row():
-        mic_audio = gr.Audio(type="filepath", label="🎤 Record Voice", interactive=True)
-        upload_file = gr.File(label="📎 Upload File")
-        upload_img = gr.Image(type="filepath", label="🖼️ Upload Image")
+        msg = gr.Textbox(placeholder="Type here or use mic...", elem_classes="message-input")
+        send_btn = gr.Button("Send")
 
-    def handle_input(message, file, image, mic_audio):
-        reply = chat(message, file, image, mic_audio)
-        return chatbot.update(chatbot.value + [[message, reply]])
+    with gr.Row():
+        record_btn = gr.Audio(source="microphone", type="filepath", label="Record Voice")
+        upload_btn = gr.File(label="Upload File", file_types=[".pdf", ".docx", ".xlsx", ".xls", ".jpg", ".jpeg", ".png"])
 
-    send_btn.click(handle_input, inputs=[txt, upload_file, upload_img, mic_audio], outputs=chatbot)
-    txt.submit(handle_input, inputs=[txt, upload_file, upload_img, mic_audio], outputs=chatbot)
+    # Event functions
+    send_btn.click(fn=chatbot_response, inputs=[msg, chatbot], outputs=[msg, chatbot])
+    record_btn.change(fn=transcribe_audio, inputs=record_btn, outputs=msg)
+    upload_btn.change(fn=extract_text_from_file, inputs=upload_btn, outputs=msg)
 
 demo.launch()