On Top of Pasketti: Children’s Speech Recognition Challenge - Word Track

In the Word Track of the On Top of Pasketti: Children's Speech Recognition Challenge, your goal is to build a model that predicts the words spoken by children in short audio clips. Word-level automatic speech recognition (ASR) recovers the intended lexical content of speech and underpins applications such as classroom transcription, voice-driven educational tools, and accessible interfaces for young learners. Reliable word-level ASR is essential for real-world educational and accessibility use cases, yet remains particularly challenging for children's speech. This track brings together well-labeled, representative children's speech data to encourage models that generalize across ages, dialects, recording environments, and real-world classroom conditions. Data for this challenge are assembled from over a dozen data sources collected under different protocols, with manual corrections and annotations added by our team. These sources have been harmonized to a common schema and released as two distinct training corpora that share the same structure but contain different data, and are hosted in separate locations for participant access.