A FiLM-conditioned transformer masker on DAC codec latents performs text-guided sound separation with claimed efficiency, but the main comparison against AudioSep is confounded by asymmetric input processing.
A simple dynamic learning rate tuning algorithm for automated training of dnns
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
CodecSep: Prompt-Driven Universal Sound Separation on Neural Audio Codec Latents
A FiLM-conditioned transformer masker on DAC codec latents performs text-guided sound separation with claimed efficiency, but the main comparison against AudioSep is confounded by asymmetric input processing.