Update app.py

app.py

@@ -3,6 +3,7 @@ import gradio as gr
 import json, re
 import openai
 from datetime import datetime
+from pydub import AudioSegment
 import pdfplumber, docx, pandas as pd
 from PIL import Image
 
@@ -60,22 +61,30 @@ def load_chat(name):
         with open(filename, "r", encoding="utf-8") as f:
             oai_history = json.load(f)
         chatbot_ui = []
-        for m in oai_history:
-            if m["role"] in ["user", "assistant"]:
-                chatbot_ui.append({"role": m["role"], "content": m["content"]})
+        for i in range(0, len(oai_history)):
+            if oai_history[i]["role"] == "user":
+                chatbot_ui.append({"role": "user", "content": oai_history[i]["content"]})
+            elif oai_history[i]["role"] == "assistant":
+                chatbot_ui.append({"role": "assistant", "content": oai_history[i]["content"]})
         return chatbot_ui, {"oai_history": oai_history, "chatbot_ui": chatbot_ui}, f"✅ Loaded {name}"
     except Exception as e:
         return [], {"oai_history": [], "chatbot_ui": []}, f"❌ Could not load {name}: {e}"
 
 # Transcription
+
 def transcribe_audio(file):
-    if not file: return ""
-    audio_file = open(file, "rb")
-    transcript = openai.Audio.transcribe("whisper-1", audio_file)
-    return transcript["text"]
+    if not file:
+        return ""
+    try:
+        with open(file, "rb") as audio_file:
+            transcript = openai.Audio.transcribe("whisper-1", audio_file)
+        return transcript["text"]
+    except Exception as e:
+        return f"❌ Transcription error: {e}"
+
+# File processor – now sends file content to model for follow-up conversation
 
-# File processor
-def process_file(file):
+def file_to_message(file, state):
     if file is None:
         return ""
     filename = file.name
@@ -91,24 +100,19 @@ def process_file(file):
             df = pd.read_csv(file.name) if ext == "csv" else pd.read_excel(file.name)
             text = df.to_string()
         elif ext in ["png", "jpg", "jpeg"]:
-            Image.open(file.name)  # just validate
-            text = "Image uploaded. Please describe what you want to know."
+            text = "[Image uploaded: please describe what you'd like me to do with it.]"
         else:
             text = file.read().decode("utf-8")
 
         if len(text) > 8000:
             text = text[:8000]
 
-        response = openai.ChatCompletion.create(
-            model="llama3-70b-8192",
-            messages=[
-                {"role": "system", "content": "Summarize or explain the content uploaded."},
-                {"role": "user", "content": text}
-            ]
-        )
-        return response["choices"][0]["message"]["content"]
+        state["oai_history"].append({"role": "user", "content": f"File uploaded:\n{text}"})
+        state["chatbot_ui"].append({"role": "user", "content": f"File uploaded:\n{text}"})
+
+        return text, state
     except Exception as e:
-        return f"❌ Error processing file: {e}"
+        return f"❌ Error processing file: {e}", state
 
 # UI
 with gr.Blocks(css="""
@@ -130,12 +134,12 @@ textarea, input[type='text'] { background: #f0f0f0; border-radius: 8px; }
         send_btn = gr.Button("🚀", scale=1)
 
     with gr.Row():
-        mic_audio = gr.Audio(type="filepath", label="🎙️ Record Voice")
+        mic_audio = gr.Audio(source="microphone", type="filepath", label="🎙️ Record Voice")
         mic_audio.change(transcribe_audio, [mic_audio], [chat_input])
 
     with gr.Row():
         file_upload = gr.File(label="📎 Upload file", file_types=[".pdf", ".docx", ".txt", ".csv", ".xlsx", ".jpg", ".png"])
-        process_btn = gr.Button("🔍 Analyze File")
+        file_upload.change(file_to_message, [file_upload, state], [chat_input, state])
 
     with gr.Row():
         new_btn = gr.Button("🆕 New Chat")
@@ -148,8 +152,6 @@ textarea, input[type='text'] { background: #f0f0f0; border-radius: 8px; }
     send_btn.click(chat_with_groq, [chat_input, state], [chat_input, chatbot, state])
     chat_input.submit(chat_with_groq, [chat_input, state], [chat_input, chatbot, state])
 
-    process_btn.click(process_file, [file_upload], [chat_input])
-
     new_btn.click(lambda: ("", [], {"oai_history": [], "chatbot_ui": []}), [], [chat_input, chatbot, state])
     save_btn.click(save_session, [state], [status])
     save_btn.click(lambda: gr.update(choices=list_saved_files()), [], [dropdown])