Towards Streaming Target Speaker Extraction via Chunk-wise Interleaved Splicing of Autoregressive Language Model

While generative models have set new benchmarks for Target Speaker Extraction (TSE), their inherent reliance on global context precludes deployment in real-time applications. Direct adaptation to streaming scenarios often leads to catastrophic inference performance degradation due to the severe mismatch between training and streaming inference. To bridge this gap, we present the first autoregressive (AR) models tailored for streaming TSE. Our approach introduces a Chunk-wise Interleaved Splicing Paradigm that ensures highly efficient and stable streaming inference. To ensure the coherence between the extracted speech segments, we design a historical context refinement mechanism that mitigates boundary discontinuities by leveraging historical information. Experiments on Libri2Mix show that while AR generative baseline exhibits performance degradation at low latencies, our approach maintains 100% stability and superior intelligibility. Furthermore, our streaming results are comparable to or even surpass offline baselines. Additionally, our model achieves a Real-Time-Factor (RTF) of 0.248 on consumer-level GPUs. This work provides empirical evidence that AR generative backbones are viable for latency-sensitive applications through the Chunk-wise Interleaved Splicing Paradigm.

VoiceMark: Zero-Shot Voice Cloning-Resistant Watermarking Approach Leveraging Speaker-Specific Latents

Voice cloning (VC)-resistant watermarking is an emerging technique for tracing and preventing unauthorized cloning. Existing methods effectively trace traditional VC models by training them on watermarked audio but fail in zero-shot VC scenarios, where models synthesize audio from an audio prompt without training. To address this, we propose VoiceMark, the first zero-shot VC-resistant watermarking method that leverages speaker-specific latents as the watermark carrier, allowing the watermark to transfer through the zero-shot VC process into the synthesized audio. Additionally, we introduce VC-simulated augmentations and VAD-based loss to enhance robustness against distortions. Experiments on multiple zero-shot VC models demonstrate that VoiceMark achieves over 95% accuracy in watermark detection after zero-shot VC synthesis, significantly outperforming existing methods, which only reach around 50%. See our code and demos at: https://huggingface.co/spaces/haiyunli/VoiceMark

Iterative Data Augmentation with Large Language Models for Aspect-based Sentiment Analysis

Aspect-based Sentiment Analysis (ABSA) is an important sentiment analysis task, which aims to determine the sentiment polarity towards an aspect in a sentence. Due to the expensive and limited labeled data, data augmentation (DA) has become the standard for improving the performance of ABSA. However, current DA methods usually have some shortcomings: 1) poor fluency and coherence, 2) lack of diversity of generated data, and 3) reliance on some existing labeled data, hindering its applications in real-world scenarios. In response to these problems, we propose a systematic Iterative Data augmentation framework, namely IterD, to boost the performance of ABSA. The core of IterD is to leverage the powerful ability of large language models (LLMs) to iteratively generate more fluent and diverse synthetic labeled data, starting from an unsupervised sentence corpus. Extensive experiments on 4 widely-used ABSA benchmarks show that IterD brings consistent and significant performance gains among 5 baseline ABSA models. More encouragingly, the synthetic data generated by IterD can achieve comparable or even better performance against the manually annotated data.