“The latest advanced models, such as Anthropic's Opus models and recent GPT versions, are highly effective at processing unstructured and nonlinear spoken input.”