A new model generates first-order ambisonics spatial audio directly from silent video and camera direction, with a new 102K-clip dataset and spatial evaluation metrics.
Audiolm: A language modeling approach to audio generation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
ViSAGe: Video-to-Spatial Audio Generation
A new model generates first-order ambisonics spatial audio directly from silent video and camera direction, with a new 102K-clip dataset and spatial evaluation metrics.