app.py

app.py

@@ -1,6 +1,13 @@
 import streamlit as st
 from openai import OpenAI
 import time
+import os
+import re
+import pandas as pd
+import PyPDF2
+from datetime import datetime
+from pydub import AudioSegment
+from docx import Document
 
 st.set_page_config(page_title="Schlager ContractAi")
 
@@ -14,10 +21,12 @@ with st.sidebar:
 # Tabs for Contract and Minutes
 tab1, tab2 = st.tabs(["Contract", "Minutes"])
 
+SUPPORTED_AUDIO_FORMATS = (".mp3", ".wav", ".m4a")
+SUPPORTED_TEXT_FORMATS = (".txt", ".docx", ".csv", ".xlsx", ".pdf")
+
 with tab1:
     st.subheader("Contract Chat")
     
-    # Initialize OpenAI client
     if OPENAI_API_KEY:
         client = OpenAI(api_key=OPENAI_API_KEY)
     else:
@@ -25,25 +34,17 @@ with tab1:
         st.stop()
 
     ASSISTANT_ID = "asst_rd9h8PfYuOmHbkvOF3RTmVfn"
-
-    # Session state for chat messages
     if "messages" not in st.session_state:
         st.session_state["messages"] = []
 
-    # Display all chat messages
     for message in st.session_state.messages:
         role, content = message["role"], message["content"]
-        if role == "assistant" and "|" in content:  # Detect table formatting
-            st.chat_message(role).write(content, unsafe_allow_html=True)
-        else:
-            st.chat_message(role).write(content)
+        st.chat_message(role).write(content)
 
-    # User input
     if prompt := st.chat_input():
         st.session_state.messages.append({"role": "user", "content": prompt})
         st.chat_message("user").write(prompt)
 
-        # Send message to OpenAI Assistant
         try:
             thread = client.beta.threads.create()
             thread_id = thread.id
@@ -66,26 +67,58 @@ with tab1:
 
             messages = client.beta.threads.messages.list(thread_id=thread_id)
             assistant_message = messages.data[0].content[0].text.value
-
-            # Detect table format and use Markdown
-            if "|" in assistant_message and "---" in assistant_message:
-                formatted_message = f"``\n{assistant_message}\n```"
-                st.chat_message("assistant").write(formatted_message, unsafe_allow_html=True)
-            else:
-                st.chat_message("assistant").write(assistant_message)
-
+            st.chat_message("assistant").write(assistant_message)
             st.session_state.messages.append({"role": "assistant", "content": assistant_message})
-
         except Exception as e:
             st.error(f"Error: {str(e)}")
 
 with tab2:
     st.subheader("Minutes")
-    st.write("Hello World")
-    
-    uploaded_files = st.file_uploader("Upload meeting minutes (PDF/DOCX)", type=["pdf", "docx"], accept_multiple_files=True)
+    uploaded_files = st.file_uploader("Upload meeting minutes (PDF/DOCX/Audio)", type=["pdf", "docx", "mp3", "wav", "m4a"], accept_multiple_files=True)
     
     if uploaded_files:
         st.write("### Uploaded Files:")
         for uploaded_file in uploaded_files:
             st.write(f"- {uploaded_file.name}")
+        
+        combined_text = ""
+        for uploaded_file in uploaded_files:
+            if uploaded_file.name.lower().endswith(SUPPORTED_AUDIO_FORMATS):
+                audio = AudioSegment.from_file(uploaded_file)
+                audio.export("temp_audio.mp3", format="mp3")
+                with open("temp_audio.mp3", "rb") as audio_file:
+                    transcription = client.audio.transcriptions.create(
+                        model="whisper-1",
+                        file=audio_file
+                    )
+                combined_text += transcription.text + "\n"
+            else:
+                if uploaded_file.name.endswith(".docx"):
+                    doc = Document(uploaded_file)
+                    combined_text += "\n".join([para.text for para in doc.paragraphs])
+                elif uploaded_file.name.endswith(".pdf"):
+                    pdf_reader = PyPDF2.PdfReader(uploaded_file)
+                    combined_text += "\n".join([page.extract_text() for page in pdf_reader.pages if page.extract_text()])
+        
+        if combined_text:
+            st.write("### Transcribed and Extracted Text:")
+            st.text_area("Meeting Transcript", combined_text, height=300)
+            
+            if st.button("Generate Meeting Minutes"):
+                response = client.chat.completions.create(
+                    model="gpt-4-turbo",
+                    messages=[
+                        {"role": "system", "content": "You are an AI assistant that generates professional meeting minutes."},
+                        {"role": "user", "content": f"Summarize the following into structured meeting minutes:\n{combined_text}"}
+                    ]
+                )
+                minutes = response.choices[0].message.content
+                st.write("### Meeting Minutes:")
+                st.text_area("Generated Minutes", minutes, height=300)
+
+                date_stamp = datetime.now().strftime("%Y-%m-%d")
+                file_name = f"Minutes_{date_stamp}.docx"
+                doc = Document()
+                doc.add_paragraph(minutes)
+                doc.save(file_name)
+                st.success(f"Meeting minutes saved as {file_name}")