Added Transcription and Webpage Tools

agent_tools.py

@@ -1,6 +1,9 @@
+from bs4 import BeautifulSoup
 from langchain_community.document_loaders import TextLoader, PyPDFLoader, Docx2txtLoader, CSVLoader
 from langchain_text_splitters import RecursiveCharacterTextSplitter
 from langchain_community.retrievers import BM25Retriever
+import requests
+import whisper
 
 @tool
 def addition_tool(a:float, b:float)->float:
@@ -110,3 +113,37 @@ class RagTool(Tool):
         retriever = BM25Retriever.from_documents(docs)
         results = retriever.get_relevant_documents(query)
         return "\n\n".join([doc.page_content for doc in results[:3]]) if results else "No matching information found"
+
+
+# WebPage Tool
+@tool
+def visit_webpage_tool(url:str)->str:
+  """
+  Takes a url of webpage and returns its content.
+  Args:
+    url: The url of the webpage.
+  """
+  try:
+    # Make a GET request to fetch the raw HTML content
+    html_content = requests.get(url).text
+
+    # Parse the html content
+    soup = BeautifulSoup(html_content, 'html')
+    return soup.text # print the parsed data of html
+  except:
+    return "Could not get information from webpage"
+
+@tool
+def transcriber_tool(file_path:str)->str:
+  """
+  Takes the file path of an audio file and returns the transcribed text
+  Args:
+    file_path: File path of audio file
+  """
+  model = whisper.load_model("base")  # You can try "small" or "medium" for better accuracy
+
+  # Transcribe the audio
+  result = model.transcribe(file_path)["text"]
+
+  # Print the text
+  return f"The result of the transcription is: {result}"