arXiv subjects

Lei Xie

Publications and source records attributed to Lei Xie.

At least 217 records · Page 12Linked to original sources

arXiv2025

DiffAttack: Diffusion-based Timbre-reserved Adversarial Attack in Speaker Identification

Being a form of biometric identification, the security of the speaker identification (SID) system is of utmost importance. To better understand the robustness of SID systems, we aim to perform more realistic attacks in SID, which are challenging for both humans and machines to detect. In this study, we propose DiffAttack, a novel timbre-reserved adversarial attack approach that exploits the capability of a diffusion-based voice conversion (DiffVC) model to generate adversarial fake audio with distinct target speaker attribution. By introducing adversarial constraints into the generative process of the diffusion-based voice conversion model, we craft fake samples that effectively mislead target models while preserving speaker-wise characteristics. Specifically, inspired by the use of randomly sampled Gaussian noise in conventional adversarial attacks and diffusion processes, we incorporate adversarial constraints into the reverse diffusion process. These constraints subtly guide the reverse diffusion process toward aligning with the target speaker distribution. Our experiments on the LibriTTS dataset indicate that DiffAttack significantly improves the attack success rate compared to vanilla DiffVC and other methods. Moreover, objective and subjective evaluations demonstrate that introducing adversarial constraints does not compromise the speech quality generated by the DiffVC model.

cs.SD↗

arXiv2025

ZSVC: Zero-shot Style Voice Conversion with Disentangled Latent Diffusion Models and Adversarial Training

Style voice conversion aims to transform the speaking style of source speech into a desired style while keeping the original speaker's identity. However, previous style voice conversion approaches primarily focus on well-defined domains such as emotional aspects, limiting their practical applications. In this study, we present ZSVC, a novel Zero-shot Style Voice Conversion approach that utilizes a speech codec and a latent diffusion model with speech prompting mechanism to facilitate in-context learning for speaking style conversion. To disentangle speaking style and speaker timbre, we introduce information bottleneck to filter speaking style in the source speech and employ Uncertainty Modeling Adaptive Instance Normalization (UMAdaIN) to perturb the speaker timbre in the style prompt. Moreover, we propose a novel adversarial training strategy to enhance in-context learning and improve style similarity. Experiments conducted on 44,000 hours of speech data demonstrate the superior performance of ZSVC in generating speech with diverse speaking styles in zero-shot scenarios.

eess.AS↗

arXiv2025

HDMoLE: Mixture of LoRA Experts with Hierarchical Routing and Dynamic Thresholds for Fine-Tuning LLM-based ASR Models

Recent advancements in integrating Large Language Models (LLM) with automatic speech recognition (ASR) have performed remarkably in general domains. While supervised fine-tuning (SFT) of all model parameters is often employed to adapt pre-trained LLM-based ASR models to specific domains, it imposes high computational costs and notably reduces their performance in general domains. In this paper, we propose a novel parameter-efficient multi-domain fine-tuning method for adapting pre-trained LLM-based ASR models to multi-accent domains without catastrophic forgetting named \textit{HDMoLE}, which leverages hierarchical routing and dynamic thresholds based on combining low-rank adaptation (LoRA) with the mixer of experts (MoE) and can be generalized to any linear layer. Hierarchical routing establishes a clear correspondence between LoRA experts and accent domains, improving cross-domain collaboration among the LoRA experts. Unlike the static Top-K strategy for activating LoRA experts, dynamic thresholds can adaptively activate varying numbers of LoRA experts at each MoE layer. Experiments on the multi-accent and standard Mandarin datasets demonstrate the efficacy of HDMoLE. Applying HDMoLE to an LLM-based ASR model projector module achieves similar performance to full fine-tuning in the target multi-accent domains while using only 9.6% of the trainable parameters required for full fine-tuning and minimal degradation in the source general domain.

cs.SD↗

arXiv2024

YingSound: Video-Guided Sound Effects Generation with Multi-modal Chain-of-Thought Controls

Generating sound effects for product-level videos, where only a small amount of labeled data is available for diverse scenes, requires the production of high-quality sounds in few-shot settings. To tackle the challenge of limited labeled data in real-world scenes, we introduce YingSound, a foundation model designed for video-guided sound generation that supports high-quality audio generation in few-shot settings. Specifically, YingSound consists of two major modules. The first module uses a conditional flow matching transformer to achieve effective semantic alignment in sound generation across audio and visual modalities. This module aims to build a learnable audio-visual aggregator (AVA) that integrates high-resolution visual features with corresponding audio features at multiple stages. The second module is developed with a proposed multi-modal visual-audio chain-of-thought (CoT) approach to generate finer sound effects in few-shot settings. Finally, an industry-standard video-to-audio (V2A) dataset that encompasses various real-world scenarios is presented. We show that YingSound effectively generates high-quality synchronized sounds across diverse conditional inputs through automated evaluations and human studies. Project Page: \url{https://giantailab.github.io/yingsound/}

cs.SD↗

arXiv2024

StableVC: Style Controllable Zero-Shot Voice Conversion with Conditional Flow Matching

Zero-shot voice conversion (VC) aims to transfer the timbre from the source speaker to an arbitrary unseen speaker while preserving the original linguistic content. Despite recent advancements in zero-shot VC using language model-based or diffusion-based approaches, several challenges remain: 1) current approaches primarily focus on adapting timbre from unseen speakers and are unable to transfer style and timbre to different unseen speakers independently; 2) these approaches often suffer from slower inference speeds due to the autoregressive modeling methods or the need for numerous sampling steps; 3) the quality and similarity of the converted samples are still not fully satisfactory. To address these challenges, we propose a style controllable zero-shot VC approach named StableVC, which aims to transfer timbre and style from source speech to different unseen target speakers. Specifically, we decompose speech into linguistic content, timbre, and style, and then employ a conditional flow matching module to reconstruct the high-quality mel-spectrogram based on these decomposed features. To effectively capture timbre and style in a zero-shot manner, we introduce a novel dual attention mechanism with an adaptive gate, rather than using conventional feature concatenation. With this non-autoregressive design, StableVC can efficiently capture the intricate timbre and style from different unseen speakers and generate high-quality speech significantly faster than real-time. Experiments demonstrate that our proposed StableVC outperforms state-of-the-art baseline systems in zero-shot VC and achieves flexible control over timbre and style from different unseen speakers. Moreover, StableVC offers approximately 25x and 1.65x faster sampling compared to autoregressive and diffusion-based baselines.

eess.AS↗

arXiv2024

Freeze-Omni: A Smart and Low Latency Speech-to-speech Dialogue Model with Frozen LLM

Rapidly developing large language models (LLMs) have brought tremendous intelligent applications. Especially, the GPT-4o's excellent duplex speech interaction ability has brought impressive experience to users. Researchers have recently proposed several multi-modal LLMs in this direction that can achieve user-agent speech-to-speech conversations. This paper proposes a novel speech-text multimodal LLM architecture called Freeze-Omni. Our main contribution is that the speech input and output modalities can be easily connected to a textual LLM while keeping the LLM's parameters frozen throughout the training process. We design a three-stage training strategy for modeling both the speech input and output, enabling Freeze-Omni to obtain speech-to-speech conversation ability using text-speech paired data (such as ASR and TTS data) and only 60,000 multi-round text Q&A data on 8 GPUs. Moreover, we can effectively ensure that the intelligence of the Freeze-Omni in the speech modality is at the same level compared with that in the text modality of its backbone LLM, while achieving low latency end-to-end spoken response. In addition, we also designed a method to achieve duplex dialogue ability through multi-task training, giving Freeze-Omni a more natural style of dialogue ability between users and agents. In summary, Freeze-Omni holds great potential to conduct speech-to-speech dialogue based on a multimodal LLM under the condition of a frozen LLM, avoiding the catastrophic forgetting problem caused by limited data and training resources.

cs.SD↗

arXiv2024

SQ-Whisper: Speaker-Querying based Whisper Model for Target-Speaker ASR

Benefiting from massive and diverse data sources, speech foundation models exhibit strong generalization and knowledge transfer capabilities to a wide range of downstream tasks. However, a limitation arises from their exclusive handling of single-speaker speech input, making them ineffective in recognizing multi-speaker overlapped speech, a common occurrence in real-world scenarios. In this study, we delve into the adaptation of speech foundation models to eliminate interfering speakers from overlapping speech and perform target-speaker automatic speech recognition (TS-ASR). Initially, we utilize the Whisper model as the foundation for adaptation and conduct a thorough comparison of its integration with existing target-speaker adaptation techniques. We then propose an innovative model termed Speaker-Querying Whisper (SQ-Whisper), which employs a set number of trainable queries to capture speaker prompts from overlapping speech based on target-speaker enrollment. These prompts serve to steer the model in extracting speaker-specific features and accurately recognizing target-speaker transcriptions. Experimental results demonstrate that our approach effectively adapts the pre-trained speech foundation model to TS-ASR. Compared with the robust TS-HuBERT model, the proposed SQ-Whisper significantly improves performance, yielding up to 15% and 10% relative reductions in word error rates (WERs) on the Libri2Mix and WSJ0-2Mix datasets, respectively. With data augmentation, we establish new state-of-the-art WERs of 14.6% on the Libri2Mix Test set and 4.4% on the WSJ0-2Mix Test set. Furthermore, we evaluate our model on the real-world AMI meeting dataset, which shows consistent improvement over other adaptation methods.

eess.AS↗

arXiv2024

CoDiff-VC: A Codec-Assisted Diffusion Model for Zero-shot Voice Conversion

Zero-shot voice conversion (VC) aims to convert the original speaker's timbre to any target speaker while keeping the linguistic content. Current mainstream zero-shot voice conversion approaches depend on pre-trained recognition models to disentangle linguistic content and speaker representation. This results in a timbre residue within the decoupled linguistic content and inadequacies in speaker representation modeling. In this study, we propose CoDiff-VC, an end-to-end framework for zero-shot voice conversion that integrates a speech codec and a diffusion model to produce high-fidelity waveforms. Our approach involves employing a single-codebook codec to separate linguistic content from the source speech. To enhance content disentanglement, we introduce Mix-Style layer normalization (MSLN) to perturb the original timbre. Additionally, we incorporate a multi-scale speaker timbre modeling approach to ensure timbre consistency and improve voice detail similarity. To improve speech quality and speaker similarity, we introduce dual classifier-free guidance, providing both content and timbre guidance during the generation process. Objective and subjective experiments affirm that CoDiff-VC significantly improves speaker similarity, generating natural and higher-quality speech.

cs.SD↗

arXiv2024

Predictive Spliner: Data-Driven Overtaking in Autonomous Racing Using Opponent Trajectory Prediction

Head-to-head racing against opponents is a challenging and emerging topic in the domain of autonomous racing. We propose Predictive Spliner, a data-driven overtaking planner that learns the behavior of opponents through Gaussian Process (GP) regression, which is then leveraged to compute viable overtaking maneuvers in future sections of the racing track. Experimentally validated on a 1:10 scale autonomous racing platform using Light Detection and Ranging (LiDAR) information to perceive the opponent, Predictive Spliner outperforms State-of-the-Art (SotA) algorithms by overtaking opponents at up to 83.1% of its own speed, being on average 8.4% faster than the previous best-performing method. Additionally, it achieves an average success rate of 84.5%, which is 47.6% higher than the previous best-performing method. The method maintains computational efficiency with a Central Processing Unit (CPU) load of 22.79% and a computation time of 8.4 ms, evaluated on a Commercial off-the-Shelf (CotS) Intel i7-1165G7, making it suitable for real-time robotic applications. These results highlight the potential of Predictive Spliner to enhance the performance and safety of autonomous racing vehicles. The code for Predictive Spliner is available at: https://github.com/ForzaETH/predictive-spliner.

cs.RO↗

arXiv2024

Learning-Based On-Track System Identification for Scaled Autonomous Racing in Under a Minute

Accurate tire modeling is crucial for optimizing autonomous racing vehicles, as state-of-the-art (SotA) model-based techniques rely on precise knowledge of the vehicle's parameters. Yet, system identification in dynamic racing conditions is challenging due to varying track and tire conditions. Traditional methods require extensive operational ranges, often impractical in racing scenarios. Machine learning (ML)-based methods, while improving performance, struggle with generalization and depend on accurate initialization. This paper introduces a novel on-track system identification algorithm, incorporating a neural network (NN) for error correction, which is then employed for traditional system identification with virtually generated data. Crucially, the process is iteratively reapplied, with tire parameters updated at each cycle, leading to notable improvements in accuracy in tests on a scaled vehicle. Experiments show that it is possible to learn a tire model without prior knowledge with only 30 seconds of driving data and 3 seconds of training time. This method demonstrates greater one-step prediction accuracy than the baseline nonlinear least squares (NLS) method under noisy conditions, achieving a 3.3x lower root mean square error (RMSE), and yields tire models with comparable accuracy to traditional steady-state system identification. Furthermore, unlike steady-state methods requiring large spaces and specific experimental setups, the proposed approach identifies tire parameters directly on a race track in dynamic racing environments.

cs.RO↗

arXiv2024

MobileMamba: Lightweight Multi-Receptive Visual Mamba Network

Previous research on lightweight models has primarily focused on CNNs and Transformer-based designs. CNNs, with their local receptive fields, struggle to capture long-range dependencies, while Transformers, despite their global modeling capabilities, are limited by quadratic computational complexity in high-resolution scenarios. Recently, state-space models have gained popularity in the visual domain due to their linear computational complexity. Despite their low FLOPs, current lightweight Mamba-based models exhibit suboptimal throughput. In this work, we propose the MobileMamba framework, which balances efficiency and performance. We design a three-stage network to enhance inference speed significantly. At a fine-grained level, we introduce the Multi-Receptive Field Feature Interaction(MRFFI) module, comprising the Long-Range Wavelet Transform-Enhanced Mamba(WTE-Mamba), Efficient Multi-Kernel Depthwise Convolution(MK-DeConv), and Eliminate Redundant Identity components. This module integrates multi-receptive field information and enhances high-frequency detail extraction. Additionally, we employ training and testing strategies to further improve performance and efficiency. MobileMamba achieves up to 83.6% on Top-1, surpassing existing state-of-the-art methods which is maximum x21 faster than LocalVim on GPU. Extensive experiments on high-resolution downstream tasks demonstrate that MobileMamba surpasses current efficient models, achieving an optimal balance between speed and accuracy.

cs.CV↗

arXiv2024

Unveiling the Potential of LLM-Based ASR on Chinese Open-Source Datasets

Large Language Models (LLMs) have demonstrated unparalleled effectiveness in various NLP tasks, and integrating LLMs with automatic speech recognition (ASR) is becoming a mainstream paradigm. Building upon this momentum, our research delves into an in-depth examination of this paradigm on a large open-source Chinese dataset. Specifically, our research aims to evaluate the impact of various configurations of speech encoders, LLMs, and projector modules in the context of the speech foundation encoder-LLM ASR paradigm. Furthermore, we introduce a three-stage training approach, expressly developed to enhance the model's ability to align auditory and textual information. The implementation of this approach, alongside the strategic integration of ASR components, enabled us to achieve the SOTA performance on the AISHELL-1, Test_Net, and Test_Meeting test sets. Our analysis presents an empirical foundation for future research in LLM-based ASR systems and offers insights into optimizing performance using Chinese datasets. We will publicly release all scripts used for data preparation, training, inference, and scoring, as well as pre-trained models and training logs to promote reproducible research.

cs.SD↗

arXiv2024

The NPU-HWC System for the ISCSLP 2024 Inspirational and Convincing Audio Generation Challenge

This paper presents the NPU-HWC system submitted to the ISCSLP 2024 Inspirational and Convincing Audio Generation Challenge 2024 (ICAGC). Our system consists of two modules: a speech generator for Track 1 and a background audio generator for Track 2. In Track 1, we employ Single-Codec to tokenize the speech into discrete tokens and use a language-model-based approach to achieve zero-shot speaking style cloning. The Single-Codec effectively decouples timbre and speaking style at the token level, reducing the acoustic modeling burden on the autoregressive language model. Additionally, we use DSPGAN to upsample 16 kHz mel-spectrograms to high-fidelity 48 kHz waveforms. In Track 2, we propose a background audio generator based on large language models (LLMs). This system produces scene-appropriate accompaniment descriptions, synthesizes background audio with Tango 2, and integrates it with the speech generated by our Track 1 system. Our submission achieves the second place and the first place in Track 1 and Track 2 respectively.

cs.SD↗

arXiv2024

The ISCSLP 2024 Conversational Voice Clone (CoVoC) Challenge: Tasks, Results and Findings

The ISCSLP 2024 Conversational Voice Clone (CoVoC) Challenge aims to benchmark and advance zero-shot spontaneous style voice cloning, particularly focusing on generating spontaneous behaviors in conversational speech. The challenge comprises two tracks: an unconstrained track without limitation on data and model usage, and a constrained track only allowing the use of constrained open-source datasets. A 100-hour high-quality conversational speech dataset is also made available with the challenge. This paper details the data, tracks, submitted systems, evaluation results, and findings.

cs.SD↗

arXiv2024

eDOC: Explainable Decoding Out-of-domain Cell Types with Evidential Learning

Single-cell RNA-seq (scRNA-seq) technology is a powerful tool for unraveling the complexity of biological systems. One of essential and fundamental tasks in scRNA-seq data analysis is Cell Type Annotation (CTA). In spite of tremendous efforts in developing machine learning methods for this problem, several challenges remains. They include identifying Out-of-Domain (OOD) cell types, quantifying the uncertainty of unseen cell type annotations, and determining interpretable cell type-specific gene drivers for an OOD case. OOD cell types are often associated with therapeutic responses and disease origins, making them critical for precision medicine and early disease diagnosis. Additionally, scRNA-seq data contains tens thousands of gene expressions. Pinpointing gene drivers underlying CTA can provide deep insight into gene regulatory mechanisms and serve as disease biomarkers. In this study, we develop a new method, eDOC, to address aforementioned challenges. eDOC leverages a transformer architecture with evidential learning to annotate In-Domain (IND) and OOD cell types as well as to highlight genes that contribute both IND cells and OOD cells in a single cell resolution. Rigorous experiments demonstrate that eDOC significantly improves the efficiency and effectiveness of OOD cell type and gene driver identification compared to other state-of-the-art methods. Our findings suggest that eDOC may provide new insights into single-cell biology.

q-bio.GN↗

arXiv2024

NTU-NPU System for Voice Privacy 2024 Challenge

In this work, we describe our submissions for the Voice Privacy Challenge 2024. Rather than proposing a novel speech anonymization system, we enhance the provided baselines to meet all required conditions and improve evaluated metrics. Specifically, we implement emotion embedding and experiment with WavLM and ECAPA2 speaker embedders for the B3 baseline. Additionally, we compare different speaker and prosody anonymization techniques. Furthermore, we introduce Mean Reversion F0 for B5, which helps to enhance privacy without a loss in utility. Finally, we explore disentanglement models, namely $β$-VAE and NaturalSpeech3 FACodec.

eess.AS↗

arXiv2024

Ideal-LLM: Integrating Dual Encoders and Language-Adapted LLM for Multilingual Speech-to-Text

Integrating audio encoders with LLMs through connectors has enabled these models to process and comprehend audio modalities, significantly enhancing speech-to-text tasks, including automatic speech recognition (ASR) and automatic speech translation (AST). However, these methods often overlook the critical aspect of language adaptation in multilingual settings, relying instead on multilingual data without adequately addressing language differences. To address this gap, we propose the Ideal-LLM model, which employs dual multilingual encoders to enrich language feature information and utilizes a language-adapted connector to target the adaptation of each language specifically. By leveraging the complementary strengths of Whisper and MMS encoders, our approach ensures richer multilingual representations. Additionally, the language-adapted connector enhances modal transformation via a language weight selector tailored for each language. Experimental results demonstrate that Ideal-LLM significantly improves ASR performance, achieving a 32.6% relative reduction in average word error rates compared to the standard speech encoder integrated with LLMs and yields an average BLEU score of 36.78 for AST task.

eess.AS↗

arXiv2024

Speaker Contrastive Learning for Source Speaker Tracing

As a form of biometric authentication technology, the security of speaker verification systems is of utmost importance. However, SV systems are inherently vulnerable to various types of attacks that can compromise their accuracy and reliability. One such attack is voice conversion, which modifies a persons speech to sound like another person by altering various vocal characteristics. This poses a significant threat to SV systems. To address this challenge, the Source Speaker Tracing Challenge in IEEE SLT2024 aims to identify the source speaker information in manipulated speech signals. Specifically, SSTC focuses on source speaker verification against voice conversion to determine whether two converted speech samples originate from the same source speaker. In this study, we propose a speaker contrastive learning-based approach for source speaker tracing to learn the latent source speaker information in converted speech. To learn a more source-speaker-related representation, we employ speaker contrastive loss during the training of the embedding extractor. This speaker contrastive loss helps identify the true source speaker embedding among several distractor speaker embeddings, enabling the embedding extractor to learn the potentially possessing source speaker information present in the converted speech. Experiments demonstrate that our proposed speaker contrastive learning system achieves the lowest EER of 16.788% on the challenge test set, securing first place in the challenge.

cs.SD↗