Open Models Almanac
Methodsmodel architecturemultimodal architecture

single mention · model architecture

Four-frame audio patches with within-patch bidirectional attention

Groups four consecutive frames into a patch whose Transformer uses bidirectional attention confined to that patch.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See multimodal architecture for the methods filed beside it.

Filed under model architecture › multimodal architecture