Brief

Researchers have developed SEAM, a new framework designed to accurately distinguish between scripted and spontaneous speech in real-time, specifically for interview guardrails. This system addresses the issue of benchmark performance being inflated by corpus-specific shortcuts rather than genuine speech style detection. SEAM incorporates uniform preprocessing, seam-aware sampling, non-speech augmentation, and a compact DistilHuBERT backbone, achieving a 0.971 ROC-AUC on an external evaluation set with 8-second windows. The framework's effectiveness is attributed to its shortcut-prevention components, demonstrating that robust detection relies on both the model architecture and careful data design and evaluation. AI

Researchers have developed SubsurfaceGen, a GPU-accelerated tool for generating realistic, field-scale 3D velocity models and seismic data. This new system addresses limitations in existing datasets for machine learning approaches to full waveform inversion (FWI). The accompanying dataset includes 4,276 2D velocity slices and seismic data from 42 diverse geological settings, designed to improve ML-based FWI for applications like carbon sequestration and hydrocarbon exploration. AI