ChildVox aggregates 17 child-centered audio datasets into a multi-task benchmark to evaluate foundation models on physiological sounds, vocalizations, syllables, and speech recognition across childhood.
InProceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 6288–6313
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ChildVox: A Speech, Audio, and Large Audio-Language Model Benchmark in Understanding and Characterizing Sound across Childhood
ChildVox aggregates 17 child-centered audio datasets into a multi-task benchmark to evaluate foundation models on physiological sounds, vocalizations, syllables, and speech recognition across childhood.