Update agent.py

agent.py

@@ -5,48 +5,17 @@ from smolagents import tool
 
 
 @tool
-def real_number_calculator(
-    a: float, b: float, operation: Literal["add", "subtract", "multiply", "divide"]
-) -> float:
-    """
-    Perform basic arithmetic operations on two real numbers.
-
-    This function acts as a simple calculator that can add, subtract, 
-    multiply, or divide two floating-point numbers.
-
-    Args:
-        a (float): The first number.
-        b (float): The second number.
-        operation (Literal["add", "subtract", "multiply", "divide"]): 
-            The arithmetic operation to perform.
-
-    Returns:
-        float: The result of the arithmetic operation.
-
-    Raises:
-        ValueError: If the operation is invalid or if division by zero is attempted.
-
-    Example:
-        >>> real_number_calculator(10, 5, "add")
-        15.0
-        >>> real_number_calculator(10, 5, "divide")
-        2.0
-    """
-
-    if operation == "add":
-        return a + b
-    elif operation == "subtract":
-        return a - b
-    elif operation == "multiply":
-        return a * b
-    elif operation == "divide":
-        if b == 0:
-            raise ValueError("Division by zero is not allowed.")
-        return a / b
-    else:
-        raise ValueError(f"Invalid operation: {operation}")
-
+def extract_text_from_audio(file_path : str) -> str:
+    """given a path to an audio file, it extract and returns the text contained in it as a string"""
 
+    import speech_recognition as sr
+    r = sr.Recognizer()
+    with sr.AudioFile(file_path) as source:
+        # listen for the data (load audio to memory)
+        audio_data = r.record(source)
+        # recognize (convert from speech to text)
+        text = r.recognize_google(audio_data)
+    return text
 
 
 class TestAgent:
@@ -75,7 +44,7 @@ class TestAgent:
         #model = InferenceClientModel("Qwen/Qwen2.5-Coder-32B-Instruct")
         # Instantiate the agent
         self.agent = CodeAgent(
-            tools=[real_number_calculator,        # homemade tool
+            tools=[extract_text_from_audio,        # homemade tool
                    DuckDuckGoSearchTool(),          # basic tools from smolagent
                    VisitWebpageTool(),
                    wikipedia_tool,                  # tool from langchain with extra parmaeters
@@ -85,8 +54,11 @@ class TestAgent:
             model=model,
             max_steps=3,
             verbosity_level=2,
-            use_structured_outputs_internally=True
+            use_structured_outputs_internally=True.   # V3. Adds structure
         )
+        # V3. add Guidance
+        prompt_for_guidance = "\n10. Provide the answer axactly as it is asked, be concise and precise\n\nNow Begin!"
+        self.agent.prompt_templates['system_prompt'] = self.agent.prompt_templates['system_prompt'] + prompt_for_guidance
 
     def __call__(self, question: str) -> str: