Inter-speaker relative cues, expressed as text differences like louder, faster, or female, guide target speech extraction and outperform random cue subsets on a new five-language dataset.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Inter-Speaker Relative Cues for Text-Guided Target Speech Extraction
Inter-speaker relative cues, expressed as text differences like louder, faster, or female, guide target speech extraction and outperform random cue subsets on a new five-language dataset.