In September 2025, Tina Raissi enhanced the rwth-i6/i6_core repository by developing two specialized job classes to improve Word Error Rate (WER) evaluation in speech recognition workflows. She implemented RescoreLatticeCacheJob, which enables flexible rescoring of lattice caches using configurable rescorer types and parameters, and IntersectStmCtmJob, which computes intersections between CTM and STM files to ensure accurate WER assessment on filtered corpora with matched recordings. Leveraging Python and her expertise in command line tools and data processing, Tina’s work integrated seamlessly with existing evaluation pipelines, providing greater flexibility, reliability, and efficiency for WER-focused research and development tasks.