Saved in:
Bibliographic Details
Main Authors: Dai, Wang, Politis, Archontis, Virtanen, Tuomas
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.01316
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914439038500864
author Dai, Wang
Politis, Archontis
Virtanen, Tuomas
author_facet Dai, Wang
Politis, Archontis
Virtanen, Tuomas
contents This paper investigates the use of relative cues for text-based target speech extraction (TSE). We first provide a theoretical justification for relative cues from the perspectives of human perception and label quantization, showing that relative cues preserve fine-grained distinctions that are often lost in absolute categorical representations for continuous-valued attributes. Building on this analysis, we propose a two-stage TSE framework in which a speech separation model first generates candidate sources, followed by a text-guided classifier that selects the target speaker based on embedding similarity. Within this framework, we train two separate classification models to evaluate the advantages of relative cues over independent cues in case of continuous-valued attributes, considering both classification accuracy and TSE performance. Experimental results demonstrate that (i) relative cues achieve higher overall classification accuracy and improved TSE performance compared with independent cues; (ii) the proposed two-stage framework substantially outperforms single-stage text-conditioned extraction methods on both signal-level and objective perceptual metrics; and (iii) several relative cues, including language, loudness, distance, temporal order, speaking duration, random cues, and all cues, can even surpass the performance of an enrollment-audio-based TSE system. Further analysis reveals notable differences in discriminative power across cue types, providing insights into the effectiveness of different relative cues for TSE.
format Preprint
id arxiv_https___arxiv_org_abs_2603_01316
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Inter-Speaker Relative Cues for Two-Stage Text-Guided Target Speech Extraction
Dai, Wang
Politis, Archontis
Virtanen, Tuomas
Audio and Speech Processing
This paper investigates the use of relative cues for text-based target speech extraction (TSE). We first provide a theoretical justification for relative cues from the perspectives of human perception and label quantization, showing that relative cues preserve fine-grained distinctions that are often lost in absolute categorical representations for continuous-valued attributes. Building on this analysis, we propose a two-stage TSE framework in which a speech separation model first generates candidate sources, followed by a text-guided classifier that selects the target speaker based on embedding similarity. Within this framework, we train two separate classification models to evaluate the advantages of relative cues over independent cues in case of continuous-valued attributes, considering both classification accuracy and TSE performance. Experimental results demonstrate that (i) relative cues achieve higher overall classification accuracy and improved TSE performance compared with independent cues; (ii) the proposed two-stage framework substantially outperforms single-stage text-conditioned extraction methods on both signal-level and objective perceptual metrics; and (iii) several relative cues, including language, loudness, distance, temporal order, speaking duration, random cues, and all cues, can even surpass the performance of an enrollment-audio-based TSE system. Further analysis reveals notable differences in discriminative power across cue types, providing insights into the effectiveness of different relative cues for TSE.
title Inter-Speaker Relative Cues for Two-Stage Text-Guided Target Speech Extraction
topic Audio and Speech Processing
url https://arxiv.org/abs/2603.01316