docs: update for pipecat PR #3915

guides/learn/function-calling.mdx

@@ -288,6 +288,7 @@ llm.register_function(
     "get_current_weather",
     fetch_weather_from_api,
     cancel_on_interruption=True,  # Cancel if user interrupts (default: True)
+    timeout_secs=30.0,  # Optional: Override global timeout for this function
 )
 ```
 
@@ -311,6 +312,7 @@ async def get_current_weather(params: FunctionCallParams, location: str, format:
 llm.register_direct_function(
     get_current_weather,
     cancel_on_interruption=False,  # Don't cancel on interruption
+    timeout_secs=60.0,  # Optional: Override global timeout for this function
 )
 ```
 
@@ -320,9 +322,12 @@ llm.register_direct_function(
 
 - **`cancel_on_interruption=True`** (default): Function call is cancelled if user interrupts
 - **`cancel_on_interruption=False`**: Function call continues even if user interrupts
+- **`timeout_secs=None`** (default): Optional per-tool timeout in seconds. Overrides the global `function_call_timeout_secs` for this specific function
 
 Use `cancel_on_interruption=False` for critical operations that should complete even if the user starts speaking. Function calls are async, so you can continue the conversation while the function executes. Once the result returns, the LLM will automatically incorporate it into the conversation context. LLMs vary in terms of how well they incorporate changes to _previous_ messages, so you may need to experiment with your LLM provider to see how it handles this.
 
+Use `timeout_secs` to set a specific timeout for a function that differs from the global default. For example, you might want a longer timeout for database queries or shorter timeouts for quick lookups.
+
 ### 3. Create the Pipeline
 
 Include your LLM service in your pipeline with the registered functions:

server/utilities/service-switchers/llm-switcher.mdx

@@ -87,6 +87,10 @@ llm_switcher.register_function(
   Whether to cancel this function call when a user interruption occurs.
 </ParamField>
 
+<ParamField path="timeout_secs" type="Optional[float]" default="None">
+  Optional per-tool timeout in seconds. Overrides the global `function_call_timeout_secs` for this specific function.
+</ParamField>
+
 ## Usage Examples
 
 ### Basic LLM Switching