Replacing standard attention with gated focused linear attention makes speech separation models run faster and use less memory while keeping separation quality close to the state of the art.
Although previous meth- ods use STFT and downsampling to reduce speech sequence length, the attention module in those still has O(N 2) time complexity
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
FLASepformer: Efficient Speech Separation with Gated Focused Linear Attention Transformer
Replacing standard attention with gated focused linear attention makes speech separation models run faster and use less memory while keeping separation quality close to the state of the art.