Executable Boundary Contracts for Sound Event Traces

TL;DR

This paper introduces executable boundary contracts for sound event traces, enabling precise measurement and analysis of boundary behavior in various sound datasets without proposing a new temporal logic.

Contribution

It defines a novel executable boundary contract framework for sound event traces, including a bounded Boolean fragment and scoring methods, with implementation and evaluation on multiple sound datasets.

Findings

Union activity can hide typed boundary failure.

External DCASE outputs provide a class indexed challenge level reference.

Standard scores and contract coordinates show interpretable disagreements.

Abstract

Sound event reports often compress timed boundary behavior into frame, segment, or event scores. This paper defines executable boundary contracts for finite sound event traces. The frame fragment is a bounded Boolean fragment embeddable in STL after grid projection. The event layer adds declared interval matching, duration clauses, fragmentation clauses, and obligation restricted vector scoring. The aim is measurement, not a new general temporal logic and not a challenge leaderboard. The artifact evaluates controlled Mini LibriSpeech seeded scenes, MAESTRO Real soundscapes, frozen pretrained timing probes, and an official DCASE 2024 Task 4 baseline track. Across these tracks, standard scores and contract coordinates disagree in interpretable ways. The strongest real corpus finding is that union activity can hide typed boundary failure, while external DCASE outputs provide a class indexed challenge level reference. Code, generated tables, manifests, and Lean checks for the finite frame core are supplied as ancillary material.