Based on what I learned in that interview, there are more classifications of sounds that the product detects. Only a subset of those classifications are surfaced to the customer as part of the product.