Adding a linguistic-constraint loss from pretrained speech or text models during training improves audio-visual target speaker extraction across languages and visual degradation, with no inference overhead.
Incorporating Linguistic Constraints from External Knowledge Source for Audio-Visual Target Speech Extraction
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Audio-visual target speaker extraction (AV-TSE) models primarily rely on target visual cues to isolate the target speaker's voice from others. We know that humans leverage linguistic knowledge, such as syntax and semantics, to support speech perception. Inspired by this, we explore the potential of pre-trained speech-language models (PSLMs) and pre-trained language models (PLMs) as auxiliary knowledge sources for AV-TSE. In this study, we propose incorporating the linguistic constraints from PSLMs or PLMs for the AV-TSE model as additional supervision signals. Without introducing any extra computational cost during inference, the proposed approach consistently improves speech quality and intelligibility. Furthermore, we evaluate our method in multi-language settings and visual cue-impaired scenarios and show robust performance gains.
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Incorporating Linguistic Constraints from External Knowledge Source for Audio-Visual Target Speech Extraction
Adding a linguistic-constraint loss from pretrained speech or text models during training improves audio-visual target speaker extraction across languages and visual degradation, with no inference overhead.