arXiv subjects

Zexin Cai

Publications and source records attributed to Zexin Cai.

At least 19 recordsLinked to original sources

arXiv2026

Brain2Speech-Net: Intelligible, Real-Time Brain-to-Speech Synthesis Without Text Decoding

The loss of speech limits communication for individuals with paralysis. Restoring speech by synthesizing it directly from neural activity is challenging: intracortical data are scarce and lack aligned targets, so most systems rely on cascaded neural-to-text-to-speech pipelines that add latency and propagate errors. We present Brain2Speech-Net, among the first single-stage frameworks to remain intelligible under limited data while removing intermediate text decoding. A differentiable phoneme bottleneck preserves linguistic structure without explicit text decoding. A lightweight deep-HMM aligner then maps this bottleneck to contextual phoneme representations in a TTS latent space. It learns monotonic alignment between neural recordings and phoneme segments without frame-level supervision, inheriting strong acoustic priors for data-efficient training. On an intracortical dataset, Brain2Speech-Net achieves strong intelligibility in objective and listening tests while running faster than real time. Unlike cascaded systems that incur high latency and direct speech-unit models that lack intelligibility, it delivers both intelligible and real-time speech.

eess.AS↗

arXiv2026

DiffAnon: Diffusion-based Prosody Control for Voice Anonymization

To preserve or not to preserve prosody is a central question in voice anonymization. Prosody conveys meaning and affect, yet is tightly coupled with speaker identity. Existing methods either discard prosody for privacy or lack a principled mechanism to control the utility-privacy trade-off, operating at fixed design points. We propose DiffAnon, a diffusion-based anonymization method with classifier-free guidance (CFG) that provides explicit, continuous inference-time control over prosody preservation. DiffAnon refines acoustic detail over semantic embeddings of an RVQ codec, enabling smooth interpolation between anonymization strength and prosodic fidelity within a single model. To the best of our knowledge, it is the first voice anonymization framework to provide structured, interpolatable inference-time prosody control. Experiments demonstrate structured trade-off behavior, achieving strong utility while maintaining competitive privacy across controllable operating points.

eess.AS↗

arXiv2026

Content Anonymization for Privacy in Long-form Audio

Voice anonymization techniques have been found to successfully obscure a speaker's acoustic identity in short, isolated utterances in benchmarks such as the VoicePrivacy Challenge. In practice, however, utterances seldom occur in isolation: long-form audio is commonplace in domains such as interviews, phone calls, and meetings. In these cases, many utterances from the same speaker are available, which pose a significantly greater privacy risk: given multiple utterances from the same speaker, an attacker could exploit an individual's vocabulary, syntax, and turns of phrase to re-identify them, even when their voice is completely disguised. To address this risk, we propose a new approach that performs a contextual rewriting of the transcripts in an ASR-TTS pipeline to eliminate speaker-specific style while preserving meaning. We present results in a long-form telephone conversation setting demonstrating the effectiveness of a content-based attack on voice-anonymized speech. Then we show how the proposed content-based anonymization methods can mitigate this risk while preserving speech utility. Overall, we find that paraphrasing is an effective defense against content-based attacks and recommend that stakeholders adopt this step to ensure anonymity in long-form audio.

cs.SD↗

arXiv2026

Universal Speech Content Factorization

We propose Universal Speech Content Factorization (USCF), a simple and invertible linear method for extracting a low-rank speech representation in which speaker timbre is suppressed while phonetic content is preserved. USCF extends Speech Content Factorization, a closed-set voice conversion (VC) method, to an open-set setting by learning a universal speech-to-content mapping via least-squares optimization and deriving speaker-specific transformations from only a few seconds of target speech. We show through embedding analysis that USCF effectively removes speaker-dependent variation. As a zero-shot VC system, USCF achieves competitive intelligibility, naturalness, and speaker similarity compared to methods that require substantially more target-speaker data or additional neural training. Finally, we demonstrate that as a training-efficient timbre-disentangled speech feature, USCF features can serve as the acoustic representation for training timbre-prompted text-to-speech models. Speech samples and code are publicly available.

eess.AS↗

arXiv2026

Integrated Spoofing-Robust Automatic Speaker Verification via a Three-Class Formulation and LLR

Spoofing-robust automatic speaker verification (SASV) aims to integrate automatic speaker verification (ASV) and countermeasure (CM). A popular solution is fusion of independent ASV and CM scores. To better modeling SASV, some frameworks integrate ASV and CM within a single network. However, these solutions are typically bi-encoder based, offer limited interpretability, and cannot be readily adapted to new evaluation parameters without retraining. Based on this, we propose a unified end-to-end framework via a three-class formulation that enables log-likelihood ratio (LLR) inference from class logits for a more interpretable decision pipeline. Experiments show comparable performance to existing methods on ASVSpoof5 and better results on SpoofCeleb. The visualization and analysis also prove that the three-class reformulation provides more interpretability.

eess.AS↗

arXiv2026

Can LLMs Help Localize Fake Words in Partially Fake Speech?

Large language models (LLMs), trained on large-scale text, have recently attracted significant attention for their strong performance across many tasks. Motivated by this, we investigate whether a text-trained LLM can help localize fake words in partially fake speech, where only specific words within a speech are edited. We build a speech LLM to perform fake word localization via next token prediction. Experiments and analyses on AV-Deepfake1M and PartialEdit indicates that the model frequently leverages editing-style pattern learned from the training data, particularly word-level polarity substitutions for those two databases we discussed, as cues for localizing fake words. Although such particular patterns provide useful information in an in-domain scenario, how to avoid over-reliance on such particular pattern and improve generalization to unseen editing styles remains an open question.

eess.AS↗

arXiv2025

GenVC: Self-Supervised Zero-Shot Voice Conversion

Most current zero-shot voice conversion methods rely on externally supervised components, particularly speaker encoders, for training. To explore alternatives that eliminate this dependency, this paper introduces GenVC, a novel framework that disentangles speaker identity and linguistic content from speech signals in a self-supervised manner. GenVC leverages speech tokenizers and an autoregressive, Transformer-based language model as its backbone for speech generation. This design supports large-scale training while enhancing both source speaker privacy protection and target speaker cloning fidelity. Experimental results demonstrate that GenVC achieves notably higher speaker similarity, with naturalness on par with leading zero-shot approaches. Moreover, due to its autoregressive formulation, GenVC introduces flexibility in temporal alignment, reducing the preservation of source prosody and speaker-specific traits, and making it highly effective for voice anonymization.

eess.AS↗

arXiv2025

Rapidly Adapting to New Voice Spoofing: Few-Shot Detection of Synthesized Speech Under Distribution Shifts

We address the challenge of detecting synthesized speech under distribution shifts -- arising from unseen synthesis methods, speakers, languages, or audio conditions -- relative to the training data. Few-shot learning methods are a promising way to tackle distribution shifts by rapidly adapting on the basis of a few in-distribution samples. We propose a self-attentive prototypical network to enable more robust few-shot adaptation. To evaluate our approach, we systematically compare the performance of traditional zero-shot detectors and the proposed few-shot detectors, carefully controlling training conditions to introduce distribution shifts at evaluation time. In conditions where distribution shifts hamper the zero-shot performance, our proposed few-shot adaptation technique can quickly adapt using as few as 10 in-distribution samples -- achieving upto 32% relative EER reduction on deepfakes in Japanese language and 20% relative reduction on ASVspoof 2021 Deepfake dataset.

eess.AS↗

arXiv2025

Scalable Controllable Accented TTS

We tackle the challenge of scaling accented TTS systems, expanding their capabilities to include much larger amounts of training data and a wider variety of accent labels, even for accents that are poorly represented or unlabeled in traditional TTS datasets. To achieve this, we employ two strategies: 1. Accent label discovery via a speech geolocation model, which automatically infers accent labels from raw speech data without relying solely on human annotation; 2. Timbre augmentation through kNN voice conversion to increase data diversity and model robustness. These strategies are validated on CommonVoice, where we fine-tune XTTS-v2 for accented TTS with accent labels discovered or enhanced using geolocation. We demonstrate that the resulting accented TTS model not only outperforms XTTS-v2 fine-tuned on self-reported accent labels in CommonVoice, but also existing accented TTS benchmarks.

eess.AS↗

arXiv2025

Self-supervised Reflective Learning through Self-distillation and Online Clustering for Speaker Representation Learning

Speaker representation learning is crucial for voice recognition systems, with recent advances in self-supervised approaches reducing dependency on labeled data. Current two-stage iterative frameworks, while effective, suffer from significant computational overhead due to repeated rounds of clustering and training. They also struggle with noisy pseudo labels that can impair model learning. This paper introduces self-supervised reflective learning (SSRL), an improved framework that addresses these limitations by enabling continuous refinement of pseudo labels during training. Through a teacher-student architecture and online clustering mechanism, SSRL eliminates the need for iterative training rounds. To handle label noise, we incorporate noisy label modeling and pseudo label queues that maintain temporal consistency. Experiments on VoxCeleb show SSRL's superiority over current two-stage iterative approaches, surpassing the performance of a 5-round method in just a single training round. Ablation studies validate the contributions of key components like noisy label modeling and pseudo label queues. Moreover, consistent improvements in pseudo labeling and the convergence of cluster counts demonstrate SSRL's effectiveness in deciphering unlabeled data. This work marks an important advancement in efficient and accurate self-supervised speaker representation learning through the novel reflective learning paradigm.

eess.AS↗

arXiv2025

ShiftySpeech: A Large-Scale Synthetic Speech Dataset with Distribution Shifts

The problem of synthetic speech detection has enjoyed considerable attention, with recent methods achieving low error rates across several established benchmarks. However, to what extent can low error rates on academic benchmarks translate to more realistic conditions? In practice, while the training set is fixed at one point in time, test-time conditions may exhibit distribution shifts relative to the training conditions, such as changes in speaker characteristics, emotional expressiveness, language and acoustic conditions, and the emergence of novel synthesis methods. Although some existing datasets target subsets of these distribution shifts, systematic analysis remains difficult due to inconsistencies between source data and synthesis systems across datasets. This difficulty is further exacerbated by the rapid development of new text-to-speech (TTS) and vocoder systems, which continually expand the diversity of synthetic speech. To enable systematic benchmarking of model performance under distribution shifts, we introduce ShiftySpeech, a large-scale benchmark comprising over 3,000 hours of synthetic speech across 7 source domains, 6 TTS systems, 12 vocoders, and 3 languages. ShiftySpeech is specifically designed to evaluate model generalization under controlled distribution shifts while ensuring broad coverage of modern synthetic speech generation techniques. It fills a key gap in current benchmarks by supporting fine-grained, controlled analysis of generalization robustness. All tested distribution shifts significantly degrade detection performance of state-of-the-art detection approaches based on self-supervised features. Overall, our findings suggest that reliance on synthetic speech detection methods in production environments should be carefully evaluated based on anticipated distribution shifts.

eess.AS↗

arXiv2024

The Database and Benchmark for the Source Speaker Tracing Challenge 2024

Voice conversion (VC) systems can transform audio to mimic another speaker's voice, thereby attacking speaker verification (SV) systems. However, ongoing studies on source speaker verification (SSV) are hindered by limited data availability and methodological constraints. This paper presents the Source Speaker Tracking Challenge (SSTC) on STL 2024, which aims to fill the gap in the database and benchmark for the SSV task. In this study, we generate a large-scale converted speech database with 16 common VC methods and train a batch of baseline systems based on the MFA-Conformer architecture. In addition, we introduced a related task called conversion method recognition, with the aim of assisting the SSV task. We expect SSTC to be a platform for advancing the development of the SSV task and provide further insights into the performance and limitations of current SV systems against VC attacks. Further details about SSTC can be found in https://sstc-challenge.github.io/.

eess.AS↗

arXiv2024

HLTCOE JHU Submission to the Voice Privacy Challenge 2024

We present a number of systems for the Voice Privacy Challenge, including voice conversion based systems such as the kNN-VC method and the WavLM voice Conversion method, and text-to-speech (TTS) based systems including Whisper-VITS. We found that while voice conversion systems better preserve emotional content, they struggle to conceal speaker identity in semi-white-box attack scenarios; conversely, TTS methods perform better at anonymization and worse at emotion preservation. Finally, we propose a random admixture system which seeks to balance out the strengths and weaknesses of the two category of systems, achieving a strong EER of over 40% while maintaining UAR at a respectable 47%.

eess.AS↗

arXiv2024

Privacy versus Emotion Preservation Trade-offs in Emotion-Preserving Speaker Anonymization

Advances in speech technology now allow unprecedented access to personally identifiable information through speech. To protect such information, the differential privacy field has explored ways to anonymize speech while preserving its utility, including linguistic and paralinguistic aspects. However, anonymizing speech while maintaining emotional state remains challenging. We explore this problem in the context of the VoicePrivacy 2024 challenge. Specifically, we developed various speaker anonymization pipelines and find that approaches either excel at anonymization or preserving emotion state, but not both simultaneously. Achieving both would require an in-domain emotion recognizer. Additionally, we found that it is feasible to train a semi-effective speaker verification system using only emotion representations, demonstrating the challenge of separating these two modalities.

eess.AS↗

arXiv2023

The DKU-DUKEECE System for the Manipulation Region Location Task of ADD 2023

This paper introduces our system designed for Track 2, which focuses on locating manipulated regions, in the second Audio Deepfake Detection Challenge (ADD 2023). Our approach involves the utilization of multiple detection systems to identify splicing regions and determine their authenticity. Specifically, we train and integrate two frame-level systems: one for boundary detection and the other for deepfake detection. Additionally, we employ a third VAE model trained exclusively on genuine data to determine the authenticity of a given audio clip. Through the fusion of these three systems, our top-performing solution for the ADD challenge achieves an impressive 82.23% sentence accuracy and an F1 score of 60.66%. This results in a final ADD score of 0.6713, securing the first rank in Track 2 of ADD 2023.

eess.AS↗

arXiv2023

SIG-VC: A Speaker Information Guided Zero-shot Voice Conversion System for Both Human Beings and Machines

Nowadays, as more and more systems achieve good performance in traditional voice conversion (VC) tasks, people's attention gradually turns to VC tasks under extreme conditions. In this paper, we propose a novel method for zero-shot voice conversion. We aim to obtain intermediate representations for speaker-content disentanglement of speech to better remove speaker information and get pure content information. Accordingly, our proposed framework contains a module that removes the speaker information from the acoustic feature of the source speaker. Moreover, speaker information control is added to our system to maintain the voice cloning performance. The proposed system is evaluated by subjective and objective metrics. Results show that our proposed system significantly reduces the trade-off problem in zero-shot voice conversion, while it also manages to have high spoofing power to the speaker verification system.

cs.SD↗

arXiv2022

Waveform Boundary Detection for Partially Spoofed Audio

The present paper proposes a waveform boundary detection system for audio spoofing attacks containing partially manipulated segments. Partially spoofed/fake audio, where part of the utterance is replaced, either with synthetic or natural audio clips, has recently been reported as one scenario of audio deepfakes. As deepfakes can be a threat to social security, the detection of such spoofing audio is essential. Accordingly, we propose to address the problem with a deep learning-based frame-level detection system that can detect partially spoofed audio and locate the manipulated pieces. Our proposed method is trained and evaluated on data provided by the ADD2022 Challenge. We evaluate our detection model concerning various acoustic features and network configurations. As a result, our detection system achieves an equal error rate (EER) of 6.58% on the ADD2022 challenge test set, which is the best performance in partially spoofed audio detection systems that can locate manipulated clips.

eess.AS↗

arXiv2022

Identifying Source Speakers for Voice Conversion based Spoofing Attacks on Speaker Verification Systems

An automatic speaker verification system aims to verify the speaker identity of a speech signal. However, a voice conversion system could manipulate a person's speech signal to make it sound like another speaker's voice and deceive the speaker verification system. Most countermeasures for voice conversion-based spoofing attacks are designed to discriminate bona fide speech from spoofed speech for speaker verification systems. In this paper, we investigate the problem of source speaker identification -- inferring the identity of the source speaker given the voice converted speech. To perform source speaker identification, we simply add voice-converted speech data with the label of source speaker identity to the genuine speech dataset during speaker embedding network training. Experimental results show the feasibility of source speaker identification when training and testing with converted speeches from the same voice conversion model(s). In addition, our results demonstrate that having more converted utterances from various voice conversion model for training helps improve the source speaker identification performance on converted utterances from unseen voice conversion models.

eess.AS↗