Update app.py

app.py

@@ -1,18 +1,24 @@
 import os
+import io
 import streamlit as st
 from streamlit_chat import message
-from pdfquery import PDFQuery
 from dotenv import load_dotenv
 from gtts import gTTS
-import io
-
+from audiorecorder import audiorecorder
+from pdfquery import PDFQuery
+import whisper
+import urllib.request
+from urllib.request import urlopen
+import ssl
+import json
+ssl._create_default_https_context = ssl._create_unverified_context
 # Load environment variables from .env file
 load_dotenv()
 
 st.set_page_config(page_title="Wise Chat Labs")
 
 def display_messages():
-    st.subheader("Chat with your pdf")
+    st.subheader("Chat with your PDF")
     for i, (msg, is_user) in enumerate(st.session_state["messages"]):
         message(msg, is_user=is_user, key=str(i))
     st.session_state["thinking_spinner"] = st.empty()
@@ -25,45 +31,61 @@ def generate_tts_audio(text):
     return audio_bytes
 
 def play_audio(audio_bytes):
-    audio_bytes.seek(0)  # Ensure the pointer is at the beginning of the IO stream
+    audio_bytes.seek(0)
     st.audio(audio_bytes, format='audio/mp3', start_time=0)
 
-def process_input():
-    if st.session_state["user_input"] and len(st.session_state["user_input"].strip()) > 0:
-        user_text = st.session_state["user_input"].strip()
-        with st.session_state["thinking_spinner"], st.spinner("Thinking..."):
-            query_text = st.session_state["pdfquery"].ask(user_text)
-        
-        audio_response = generate_tts_audio(query_text)
-        play_audio(audio_response)
-        
-        st.session_state["messages"].append((user_text, True))
-        st.session_state["messages"].append((query_text, False))
+def process_input(user_text):
+    with st.session_state["thinking_spinner"], st.spinner("Thinking..."):
+        query_text = st.session_state["pdfquery"].ask(user_text)
+    
+    audio_response = generate_tts_audio(query_text)
+    play_audio(audio_response)
+    
+    st.session_state["messages"].append((user_text, True))
+    st.session_state["messages"].append((query_text, False))
 
 def load_and_ingest_document():
-    file_path = 'osha.pdf'  # PDF file is assumed to be in the same directory as this script
+    file_path = 'osha.pdf'
     if os.path.exists(file_path):
         st.session_state["pdfquery"].ingest(file_path)
     else:
-        st.error("PDF file not found. Please ensure 'data.pdf' is in the root directory.")
+        st.error("PDF file not found. Please ensure 'osha.pdf' is in the root directory.")
 
 def is_openai_api_key_set() -> bool:
-    return len(st.session_state["OPENAI_API_KEY"]) > 0
+    return len(st.session_state.get("OPENAI_API_KEY", "")) > 0
+
+def transcribe_audio_to_text(audio_file_path):
+    model = whisper.load_model("base")
+    result = model.transcribe(audio_file_path)
+    return result["text"]
 
 def main():
-    if len(st.session_state) == 0:
-        st.session_state["messages"] = []
-        st.session_state["OPENAI_API_KEY"] = os.getenv("OPENAI_API_KEY", "")
+    if 'messages' not in st.session_state:
+        st.session_state['messages'] = []
+    
+    st.header("Wise Chat Labs - Your Ultimate Chatbot")
+    display_messages()
+
+    if 'OPENAI_API_KEY' not in st.session_state:
+        st.session_state['OPENAI_API_KEY'] = os.getenv("OPENAI_API_KEY", "")
         if is_openai_api_key_set():
-            st.session_state["pdfquery"] = PDFQuery(st.session_state["OPENAI_API_KEY"])
+            st.session_state['pdfquery'] = PDFQuery(st.session_state["OPENAI_API_KEY"])
             load_and_ingest_document()
         else:
-            st.session_state["pdfquery"] = None
+            st.session_state['pdfquery'] = None
+            st.error("OpenAI API key is not set. Please check your .env file.")
 
-    st.header("Wise Chat Labs - Your Ultimate Chatbot")
+    user_input = st.text_input("Type your message here...", key="user_input")
+    if user_input:
+        process_input(user_input)
 
-    display_messages()
-    st.text_input("Message", key="user_input", disabled=not is_openai_api_key_set(), on_change=process_input)
+    st.title("Or speak your message...")
+    audio = audiorecorder("Start Talking", "Click to stop", key="audio_recorder")
+    if audio is not None and hasattr(audio, 'export'):
+        audio_file_path = "user_audio_input.wav"
+        audio.export(audio_file_path, format="wav")
+        user_text = transcribe_audio_to_text(audio_file_path)
+        process_input(user_text)
 
 if __name__ == "__main__":
     main()