Speaker-Consistent Grounding for Confusion-Resilient LLM-Based Target Speech Extraction
Authors: Hanyu Meng, Hao Shi, Zhanhong He, Qiquan Zhang, Yang Xiao, Sungwoo Cho, Haowei Lou, Haizhou Li
Organizations: UNSW Sydney, Australia · Kyoto University, Japan · University of Western Australia, Australia · University of Melbourne, Australia · KAIST AI, Republic of Korea · CUHK-Shenzhen, China
Deterministic target-speaker extraction (TSE) tends to preserve observed linguistic content but may extract the interfering speaker. In contrast, autoregressive generative TSE improves perceptual quality, but is prone to content hallucination and speaker identity drift, which can be mitigated by model grounding with the evidence from the deterministic TSE. However, grounding such a generative TSE model with speaker-confused evidence may therefore reinforce the error. To address this gap, we proposed a repair-before-grounding framework with a training-free evidence-repair stage. Conditioning-diverse candidate selection (CDCS) selects target speaker-consistent evidence from complementary frozen TSE outputs. The selected waveform can then be used by code-space grounding (CSG) to constrain LLM-based TSE decoding. On Libri2Mix, CDCS reduces wrong-speaker extraction errors from 6.87% to 0.75% on clean mixtures and from 18.95% to 10.46% under noise. With the LLM-based TSE checkpoint unchanged, CSG further reduces noisy content- and acoustic-switch rates from 12.79% and 1.68% to 9.10% and 0.51%, respectively. Despite this stronger grounding, generative TSE retains its predicted perceptual-quality advantage over direct deterministic extraction. These results show that repairing evidence identity before grounding improves the reliability of generative TSE. Audio demos and code are available at https://hanyu-meng.github.io/LLM-TSE-Grounding-Demo/.
Figures & tables
Figure 1: Overview of the proposed repair-before-grounding framework. (a) CDCS-5 selects target-speaker-consistent evidence from five frozen deterministic TSE estimates. The selected waveform is either returned directly or (b) tokenized as evidence together with mixture and target-speaker conditioning. (c) CSG uses this evidence to constrain autoregressive Qwen-TSE decoding.
Table 1: Deterministic and generative TSE under clean and noisy conditions. WER and switch rates are percentages; DNSMOS denotes P.808. BSRNN-TSE is the full-enrolment baseline. Best and second-best values per condition are bold and underlined, respectively.
Target speaker extraction (TSE) aims to recover a target speaker's speech from a mixture using a reference utterance as a cue. Most TSE systems adopt conditional auto-encoder architectures with one-step inference. Inspired by test-time scaling, we propose a training-free multi-step inference method that enables iterative refinement with a frozen pretrained model. At each step, new candidates are generated by interpolating the original mixture and the previous estimate, and the best candidate is selected for further refinement until convergence. Experiments show that, when ground-truth target speech is available, optimizing an intrusive metric (SI-SDRi) yields consistent gains across multiple evaluation metrics. Without ground truth, optimizing non-intrusive metrics (UTMOS or SpkSim) improves the corresponding metric but may hurt others. We therefore introduce joint metric optimization to balance these objectives, enabling controllable extraction preferences for practical deployment.
Zhenghai You, Ying Shi, Lantian Li +1
Beijing University of Posts and Telecommunications, China · Center for Speech and Language Technologies, BNRist, Tsinghua University, China
Target speaker extraction (TSE) aims to recover a target speaker's speech from a speech mixture using an enrollment utterance. Recent codec-based generative approaches formulate TSE as conditional speech token generation and achieve promising perceptual quality. However, most existing systems rely on unified codec tokens that entangle linguistic content, speaker characteristics, and acoustic details, forcing a single model to jointly handle target selection, content semantic, and acoustic reconstruction. In this paper, we investigate whether disentangled semantic-acoustic codec representations are better suited to TSE. We first show that simply replacing unified codec tokens with semantic-acoustic tokens is insufficient and can even degrade performance when both remain handled by a unified generative model. Motivated by this observation, we propose a task-aware discriminative-generative framework, where a discriminative branch extracts target-related semantic information, while a generative branch predicts coarse acoustic tokens and refines them under semantic guidance. This specialization separates what is said from how it sounds while coordinating both factors in a factorized codec space. Experiments on Libri2Mix show that our method reduces dWER from 0.388 to 0.118 compared with a unified-codec baseline, while maintaining competitive perceptual quality and speaker consistency. Ablation studies confirm that the gains arise from task-aware semantic-acoustic modeling rather than representation disentanglement alone.
Wang Xiang, Bang Zeng, Ming Li
Wuhan University · The Chinese University of Hong Kong, Shenzhen
Target speaker extraction (TSE) isolates a desired speaker from a multi-speaker mixture using cues that identify the target. Prior work has explored acoustic and textual cues, such as enrollment audio and different types of speaker descriptors, but typically relies on a single cue family. Yet no single cue is both broadly available and reliably discriminative. Moreover, most TSE systems adopt mask-prediction models, which are prone to audible artifacts and generalize poorly across acoustic conditions. To address these limitations, we propose PromptSep-Speaker, a generative TSE model based on latent diffusion that unifies four interchangeable cues: speaker temporal order, transcript, enrollment audio, and speaker caption. The generative formulation delivers higher separation quality and multi-modal prompting improves applicability across scenarios. To fully exploit each cue, we further introduce a frozen WavLM representation with an auxiliary alignment loss that supervises query-mixture cross-attention without extra labeling. Across four evaluation sets and a listening study, PromptSep-Speaker achieves state-of-the-art performance over TSE and universal source separation baselines, improving separation accuracy and quality across cue modalities.
Yutong Wen, Ke Chen, Prem Seetharaman +5
Adobe Research, USA · Massachusetts Institute of Technology, USA