Update app.py

app.py

@@ -9,6 +9,8 @@ from langchain.prompts import PromptTemplate
 import tempfile
 from gtts import gTTS
 import os
+import docx
+from pptx import Presentation
 
 def text_to_speech(text):
     tts = gTTS(text=text, lang='en')
@@ -18,19 +20,44 @@ def text_to_speech(text):
     st.audio(temp_filename, format='audio/mp3')
     os.remove(temp_filename)
 
-def get_pdf_text(pdf_docs):
-    text=""
-    for pdf in pdf_docs:
-        pdf_reader= PdfReader(pdf)
-        for page in pdf_reader.pages:
-            text+= page.extract_text()
-    return  text
+def read_text_from_pdf(pdf_file):
+    pdf_reader = PdfReader(pdf_file)
+    text = ""
+    for page in pdf_reader.pages:
+        text += page.extract_text()
+    return text
+
+def read_text_from_docx(docx_file):
+    doc = docx.Document(docx_file)
+    text = "\n".join([paragraph.text for paragraph in doc.paragraphs])
+    return text
+
+def read_text_from_pptx(pptx_file):
+    presentation = Presentation(pptx_file)
+    text = ""
+    for slide in presentation.slides:
+        for shape in slide.shapes:
+            if hasattr(shape, "text"):
+                text += shape.text + "\n"
+    return text
+
+def get_text_from_file(file):
+    content = ""
+    if file.type == "application/pdf":
+        content = read_text_from_pdf(file)
+    elif file.type == "application/vnd.openxmlformats-officedocument.wordprocessingml.document":
+        content = read_text_from_docx(file)
+    elif file.type == "application/vnd.openxmlformats-officedocument.presentationml.presentation":
+        content = read_text_from_pptx(file)
+    elif file.type == "text/plain":
+        content = file.getvalue().decode("utf-8")
+    return content
 
 def get_text_chunks(text):
     text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
     chunks = text_splitter.split_text(text)
     return chunks
-    
+
 def get_vector_store(text_chunks, api_key):
     embeddings = HuggingFaceInferenceAPIEmbeddings(api_key=api_key, model_name="sentence-transformers/all-MiniLM-l6-v2")
     vector_store = FAISS.from_texts(text_chunks, embedding=embeddings)
@@ -62,11 +89,10 @@ def user_input(user_question, api_key):
     chain = get_conversational_chain()
 
     response = chain(
-        {"input_documents":docs, "question": user_question}
-        , return_only_outputs=True)
+        {"input_documents": docs, "question": user_question},
+        return_only_outputs=True
+    )
 
-    print(response)  # Debugging line
-    
     st.write("Replies:")
     if isinstance(response["output_text"], str):
         response_list = [response["output_text"]]
@@ -87,23 +113,25 @@ def main():
 
     with st.sidebar:
         st.title("Menu:")
-        pdf_docs = st.file_uploader("Upload your PDF Files and Click on the Submit & Process Button", accept_multiple_files=True)
+        uploaded_files = st.file_uploader("Upload your files (PDF, DOCX, PPTX, TXT)", accept_multiple_files=True)
         if st.button("Submit & Process"):
             with st.spinner("Processing..."):
-                raw_text = get_pdf_text(pdf_docs)
+                raw_text = ""
+                for file in uploaded_files:
+                    file_text = get_text_from_file(file)
+                    raw_text += file_text
                 text_chunks = get_text_chunks(raw_text)
                 get_vector_store(text_chunks, api_key)
                 st.success("Done")
 
     # Check if any document is uploaded
-    if pdf_docs:
+    if uploaded_files:
         user_question = st.text_input("Ask a question from the Docs")
 
         if user_question:
             user_input(user_question, api_key)
     else:
-        st.write("Please upload a document first to ask questions.")
+        st.write("Please upload a document (PDF, DOCX, PPTX, TXT) first to ask questions.")
 
-                
 if __name__ == "__main__":
-    main()
+    main()