Searcharxiv⌕ Search

arXiv subjects

Lei Xie

Publications and source records attributed to Lei Xie.

At least 325 records · Page 18Linked to original sources
arXiv2023

StyleS2ST: Zero-shot Style Transfer for Direct Speech-to-speech Translation

Direct speech-to-speech translation (S2ST) has gradually become popular as it has many advantages compared with cascade S2ST. However, current research mainly focuses on the accuracy of semantic translation and ignores the speech style transfer from a source language to a target language. The lack of high-fidelity expressive parallel data makes such style transfer challenging, especially in more practical zero-shot scenarios. To solve this problem, we first build a parallel corpus using a multi-lingual multi-speaker text-to-speech synthesis (TTS) system and then propose the StyleS2ST model with cross-lingual speech style transfer ability based on a style adaptor on a direct S2ST system framework. Enabling continuous style space modeling of an acoustic model through parallel corpus training and non-parallel TTS data augmentation, StyleS2ST captures cross-lingual acoustic feature mapping from the source to the target language. Experiments show that StyleS2ST achieves good style similarity and naturalness in both in-set and out-of-set zero-shot scenarios.

cs.SD↗
arXiv2023

Physics-aware Graph Neural Network for Accurate RNA 3D Structure Prediction

Biological functions of RNAs are determined by their three-dimensional (3D) structures. Thus, given the limited number of experimentally determined RNA structures, the prediction of RNA structures will facilitate elucidating RNA functions and RNA-targeted drug discovery, but remains a challenging task. In this work, we propose a Graph Neural Network (GNN)-based scoring function trained only with the atomic types and coordinates on limited solved RNA 3D structures for distinguishing accurate structural models. The proposed Physics-aware Multiplex Graph Neural Network (PaxNet) separately models the local and non-local interactions inspired by molecular mechanics. Furthermore, PaxNet contains an attention-based fusion module that learns the individual contribution of each interaction type for the final prediction. We rigorously evaluate the performance of PaxNet on two benchmarks and compare it with several state-of-the-art baselines. The results show that PaxNet significantly outperforms all the baselines overall, and demonstrate the potential of PaxNet for improving the 3D structure modeling of RNA and other macromolecules. Our code is available at https://github.com/zetayue/Physics-aware-Multiplex-GNN.

cs.LG↗
arXiv2023

Learning Universal and Robust 3D Molecular Representations with Graph Convolutional Networks

To learn accurate representations of molecules, it is essential to consider both chemical and geometric features. To encode geometric information, many descriptors have been proposed in constrained circumstances for specific types of molecules and do not have the properties to be ``robust": 1. Invariant to rotations and translations; 2. Injective when embedding molecular structures. In this work, we propose a universal and robust Directional Node Pair (DNP) descriptor based on the graph representations of 3D molecules. Our DNP descriptor is robust compared to previous ones and can be applied to multiple molecular types. To combine the DNP descriptor and chemical features in molecules, we construct the Robust Molecular Graph Convolutional Network (RoM-GCN) which is capable to take both node and edge features into consideration when generating molecule representations. We evaluate our model on protein and small molecule datasets. Our results validate the superiority of the DNP descriptor in incorporating 3D geometric information of molecules. RoM-GCN outperforms all compared baselines.

cs.LG↗
arXiv2023

Contextualized End-to-End Speech Recognition with Contextual Phrase Prediction Network

Contextual information plays a crucial role in speech recognition technologies and incorporating it into the end-to-end speech recognition models has drawn immense interest recently. However, previous deep bias methods lacked explicit supervision for bias tasks. In this study, we introduce a contextual phrase prediction network for an attention-based deep bias method. This network predicts context phrases in utterances using contextual embeddings and calculates bias loss to assist in the training of the contextualized model. Our method achieved a significant word error rate (WER) reduction across various end-to-end speech recognition models. Experiments on the LibriSpeech corpus show that our proposed model obtains a 12.1% relative WER improvement over the baseline model, and the WER of the context phrases decreases relatively by 40.5%. Moreover, by applying a context phrase filtering strategy, we also effectively eliminate the WER degradation when using a larger biasing list.

eess.AS↗
arXiv2023

The NPU-MSXF Speech-to-Speech Translation System for IWSLT 2023 Speech-to-Speech Translation Task

This paper describes the NPU-MSXF system for the IWSLT 2023 speech-to-speech translation (S2ST) task which aims to translate from English speech of multi-source to Chinese speech. The system is built in a cascaded manner consisting of automatic speech recognition (ASR), machine translation (MT), and text-to-speech (TTS). We make tremendous efforts to handle the challenging multi-source input. Specifically, to improve the robustness to multi-source speech input, we adopt various data augmentation strategies and a ROVER-based score fusion on multiple ASR model outputs. To better handle the noisy ASR transcripts, we introduce a three-stage fine-tuning strategy to improve translation accuracy. Finally, we build a TTS model with high naturalness and sound quality, which leverages a two-stage framework, using network bottleneck features as a robust intermediate representation for speaker timbre and linguistic content disentanglement. Based on the two-stage framework, pre-trained speaker embedding is leveraged as a condition to transfer the speaker timbre in the source English speech to the translated Chinese speech. Experimental results show that our system has high translation accuracy, speech naturalness, sound quality, and speaker similarity. Moreover, it shows good robustness to multi-source data.

cs.SD↗
arXiv2023

Robo-centric ESDF: A Fast and Accurate Whole-body Collision Evaluation Tool for Any-shape Robotic Planning

For letting mobile robots travel flexibly through complicated environments, increasing attention has been paid to the whole-body collision evaluation. Most existing works either opt for the conservative corridor-based methods that impose strict requirements on the corridor generation, or ESDF-based methods that suffer from high computational overhead. It is still a great challenge to achieve fast and accurate whole-body collision evaluation. In this paper, we propose a Robo-centric ESDF (RC-ESDF) that is pre-built in the robot body frame and is capable of seamlessly applied to any-shape mobile robots, even for those with non-convex shapes. RC-ESDF enjoys lazy collision evaluation, which retains only the minimum information sufficient for whole-body safety constraint and significantly speeds up trajectory optimization. Based on the analytical gradients provided by RC-ESDF, we optimize the position and rotation of robot jointly, with whole-body safety, smoothness, and dynamical feasibility taken into account. Extensive simulation and real-world experiments verified the reliability and generalizability of our method.

cs.RO↗
arXiv2023

MSW-Transformer: Multi-Scale Shifted Windows Transformer Networks for 12-Lead ECG Classification

Automatic classification of electrocardiogram (ECG) signals plays a crucial role in the early prevention and diagnosis of cardiovascular diseases. While ECG signals can be used for the diagnosis of various diseases, their pathological characteristics exhibit minimal variations, posing a challenge to automatic classification models. Existing methods primarily utilize convolutional neural networks to extract ECG signal features for classification, which may not fully capture the pathological feature differences of different diseases. Transformer networks have advantages in feature extraction for sequence data, but the complete network is complex and relies on large-scale datasets. To address these challenges, we propose a single-layer Transformer network called Multi-Scale Shifted Windows Transformer Networks (MSW-Transformer), which uses a multi-window sliding attention mechanism at different scales to capture features in different dimensions. The self-attention is restricted to non-overlapping local windows via shifted windows, and different window scales have different receptive fields. A learnable feature fusion method is then proposed to integrate features from different windows to further enhance model performance. Furthermore, we visualize the attention mechanism of the multi-window shifted mechanism to achieve better clinical interpretation in the ECG classification task. The proposed model achieves state-of-the-art performance on five classification tasks of the PTBXL-2020 12-lead ECG dataset, which includes 5 diagnostic superclasses, 23 diagnostic subclasses, 12 rhythm classes, 17 morphology classes, and 44 diagnosis classes, with average macro-F1 scores of 77.85%, 47.57%, 66.13%, 34.60%, and 34.29%, and average sample-F1 scores of 81.26%, 68.27%, 91.32%, 50.07%, and 63.19%, respectively.

eess.SP↗
arXiv2023

DCCRN-KWS: an audio bias based model for noise robust small-footprint keyword spotting

Real-world complex acoustic environments especially the ones with a low signal-to-noise ratio (SNR) will bring tremendous challenges to a keyword spotting (KWS) system. Inspired by the recent advances of neural speech enhancement and context bias in speech recognition, we propose a robust audio context bias based DCCRN-KWS model to address this challenge. We form the whole architecture as a multi-task learning framework for both denosing and keyword spotting, where the DCCRN encoder is connected with the KWS model. Helped with the denoising task, we further introduce an audio context bias module to leverage the real keyword samples and bias the network to better iscriminate keywords in noisy conditions. Feature merge and complex context linear modules are also introduced to strength such discrimination and to effectively leverage contextual information respectively. Experiments on the internal challenging dataset and the HIMIYA public dataset show that our DCCRN-KWS system is superior in performance, while ablation study demonstrates the good design of the whole model.

eess.AS↗
arXiv2023

Two Stage Contextual Word Filtering for Context bias in Unified Streaming and Non-streaming Transducer

It is difficult for an E2E ASR system to recognize words such as entities appearing infrequently in the training data. A widely used method to mitigate this issue is feeding contextual information into the acoustic model. Previous works have proven that a compact and accurate contextual list can boost the performance significantly. In this paper, we propose an efficient approach to obtain a high quality contextual list for a unified streaming/non-streaming based E2E model. Specifically, we make use of the phone-level streaming output to first filter the predefined contextual word list then fuse it into non-casual encoder and decoder to generate the final recognition results. Our approach improve the accuracy of the contextual ASR system and speed up the inference process. Experiments on two datasets demonstrates over 20% CER reduction comparing to the baseline system. Meanwhile, the RTF of our system can be stabilized within 0.15 when the size of the contextual word list grows over 6,000.

cs.SD↗
arXiv2023

PromptStyle: Controllable Style Transfer for Text-to-Speech with Natural Language Descriptions

Style transfer TTS has shown impressive performance in recent years. However, style control is often restricted to systems built on expressive speech recordings with discrete style categories. In practical situations, users may be interested in transferring style by typing text descriptions of desired styles, without the reference speech in the target style. The text-guided content generation techniques have drawn wide attention recently. In this work, we explore the possibility of controllable style transfer with natural language descriptions. To this end, we propose PromptStyle, a text prompt-guided cross-speaker style transfer system. Specifically, PromptStyle consists of an improved VITS and a cross-modal style encoder. The cross-modal style encoder constructs a shared space of stylistic and semantic representation through a two-stage training process. Experiments show that PromptStyle can achieve proper style transfer with text prompts while maintaining relatively high stability and speaker similarity. Audio samples are available in our demo page.

cs.SD↗
arXiv2023

DualVC: Dual-mode Voice Conversion using Intra-model Knowledge Distillation and Hybrid Predictive Coding

Voice conversion is an increasingly popular technology, and the growing number of real-time applications requires models with streaming conversion capabilities. Unlike typical (non-streaming) voice conversion, which can leverage the entire utterance as full context, streaming voice conversion faces significant challenges due to the missing future information, resulting in degraded intelligibility, speaker similarity, and sound quality. To address this challenge, we propose DualVC, a dual-mode neural voice conversion approach that supports both streaming and non-streaming modes using jointly trained separate network parameters. Furthermore, we propose intra-model knowledge distillation and hybrid predictive coding (HPC) to enhance the performance of streaming conversion. Additionally, we incorporate data augmentation to train a noise-robust autoregressive decoder, improving the model's performance on long-form speech conversion. Experimental results demonstrate that the proposed model outperforms the baseline models in the context of streaming voice conversion, while maintaining comparable performance to the non-streaming topline system that leverages the complete context, albeit with a latency of only 252.8 ms.

eess.AS↗
arXiv2023

Pseudo-Siamese Network based Timbre-reserved Black-box Adversarial Attack in Speaker Identification

In this study, we propose a timbre-reserved adversarial attack approach for speaker identification (SID) to not only exploit the weakness of the SID model but also preserve the timbre of the target speaker in a black-box attack setting. Particularly, we generate timbre-reserved fake audio by adding an adversarial constraint during the training of the voice conversion model. Then, we leverage a pseudo-Siamese network architecture to learn from the black-box SID model constraining both intrinsic similarity and structural similarity simultaneously. The intrinsic similarity loss is to learn an intrinsic invariance, while the structural similarity loss is to ensure that the substitute SID model shares a similar decision boundary to the fixed black-box SID model. The substitute model can be used as a proxy to generate timbre-reserved fake audio for attacking. Experimental results on the Audio Deepfake Detection (ADD) challenge dataset indicate that the attack success rate of our proposed approach yields up to 60.58% and 55.38% in the white-box and black-box scenarios, respectively, and can deceive both human beings and machines.

cs.SD↗
arXiv2023

DSPGAN: a GAN-based universal vocoder for high-fidelity TTS by time-frequency domain supervision from DSP

Recent development of neural vocoders based on the generative adversarial neural network (GAN) has shown obvious advantages of generating raw waveform conditioned on mel-spectrogram with fast inference speed and lightweight networks. Whereas, it is still challenging to train a universal neural vocoder that can synthesize high-fidelity speech from various scenarios with unseen speakers, languages, and speaking styles. In this paper, we propose DSPGAN, a GAN-based universal vocoder for high-fidelity speech synthesis by applying the time-frequency domain supervision from digital signal processing (DSP). To eliminate the mismatch problem caused by the ground-truth spectrograms in the training phase and the predicted spectrograms in the inference phase, we leverage the mel-spectrogram extracted from the waveform generated by a DSP module, rather than the predicted mel-spectrogram from the Text-to-Speech (TTS) acoustic model, as the time-frequency domain supervision to the GAN-based vocoder. We also utilize sine excitation as the time-domain supervision to improve the harmonic modeling and eliminate various artifacts of the GAN-based vocoder. Experiments show that DSPGAN significantly outperforms the compared approaches and it can generate high-fidelity speech for various TTS models trained using diverse data.

cs.SD↗
arXiv2023

Joint BS Selection, User Association, and Beamforming Design for Network Integrated Sensing and Communication

Different from conventional radar, the cellular network in the integrated sensing and communication (ISAC) system enables collaborative sensing by multiple sensing nodes, e.g., base stations (BSs). However, existing works normally assume designated BSs as the sensing nodes, and thus can't fully exploit the macro-diversity gain. In the paper, we propose a joint BS selection, user association, and beamforming design to tackle this problem. The total transmit power is minimized while guaranteeing the communication and sensing performance measured by the signal-to-interference-plus-noise ratio (SINR) for the communication users and the Cramer-Rao lower bound (CRLB) for location estimation, respectively. An alternating optimization (AO)-based algorithm is developed to solve the non-convex problem. Simulation results validate the effectiveness of the proposed algorithm and unveil the benefits brought by collaborative sensing and BS selection.

eess.SP↗
arXiv2023

Networked Sensing with AI-Empowered Interference Management: Exploiting Macro-Diversity and Array Gain in Perceptive Mobile Networks

Sensing will be an important service of future wireless networks to assist innovative applications such as autonomous driving and environment monitoring. Perceptive mobile networks (PMNs) were proposed to add sensing capability to current cellular networks. Different from traditional radar, the cellular structure of PMNs offers multiple perspectives to sense the same target, but the inherent interference between sensing and communication along with the joint processing among distributed sensing nodes (SNs) also cause big challenges for the design of PMNs. In this paper, we first propose a two-stage protocol to tackle the interference between two sub-systems. Specifically, the echoes created by communication signals, i.e., interference for sensing, are first estimated in the clutter estimation (CE) stage and then utilized for interference management in the target sensing (TS) stage. A networked sensing detector is then derived to exploit the perspectives provided by multiple SNs for sensing the same target. The macro-diversity from multiple SNs together with the array gain and the higher angular resolution from multiple receive antennas of each SN are investigated to reveal the benefit of networked sensing. Furthermore, we derive the sufficient condition to guarantee one SN's contribution is positive, based on which a SN selection algorithm is proposed. To reduce the communication workload, we propose a distributed model-driven deep-learning algorithm that utilizes partially-sampled data for CE. Simulation results confirm the benefits of networked sensing and validate the higher efficiency of the proposed CE algorithm than existing methods.

cs.IT↗
arXiv2023

Distance-based Weight Transfer from Near-field to Far-field Speaker Verification

The scarcity of labeled far-field speech is a constraint for training superior far-field speaker verification systems. Fine-tuning the model pre-trained on large-scale near-field speech substantially outperforms training from scratch. However, the fine-tuning method suffers from two limitations--catastrophic forgetting and overfitting. In this paper, we propose a weight transfer regularization(WTR) loss to constrain the distance of the weights between the pre-trained model with large-scale near-field speech and the fine-tuned model through a small number of far-field speech. With the WTR loss, the fine-tuning process takes advantage of the previously acquired discriminative ability from the large-scale near-field speech without catastrophic forgetting. Meanwhile, we use the PAC-Bayes generalization theory to analyze the generalization bound of the fine-tuned model with the WTR loss. The analysis result indicates that the WTR term makes the fine-tuned model have a tighter generalization upper bound. Moreover, we explore three kinds of norm distance for weight transfer, which are L1-norm distance, L2-norm distance and Max-norm distance. Finally, we evaluate the effectiveness of the WTR loss on VoxCeleb (pre-trained dataset) and FFSVC (fine-tuned dataset) datasets.

cs.SD↗
arXiv2023

The NPU-Elevoc Personalized Speech Enhancement System for ICASSP2023 DNS Challenge

This paper describes our NPU-Elevoc personalized speech enhancement system (NAPSE) for the 5th Deep Noise Suppression Challenge at ICASSP 2023. Based on the superior two-stage model TEA-PSE 2.0, our system particularly explores better strategy for speaker embedding fusion, optimizes the model training pipeline, and leverages adversarial training and multi-scale loss. According to the results, our system is tied for the 1st place in the headset track (track 1) and ranked 2nd in the speakerphone track (track 2).

eess.AS↗
arXiv2023

Delivering Speaking Style in Low-resource Voice Conversion with Multi-factor Constraints

Conveying the linguistic content and maintaining the source speech's speaking style, such as intonation and emotion, is essential in voice conversion (VC). However, in a low-resource situation, where only limited utterances from the target speaker are accessible, existing VC methods are hard to meet this requirement and capture the target speaker's timber. In this work, a novel VC model, referred to as MFC-StyleVC, is proposed for the low-resource VC task. Specifically, speaker timbre constraint generated by clustering method is newly proposed to guide target speaker timbre learning in different stages. Meanwhile, to prevent over-fitting to the target speaker's limited data, perceptual regularization constraints explicitly maintain model performance on specific aspects, including speaking style, linguistic content, and speech quality. Besides, a simulation mode is introduced to simulate the inference process to alleviate the mismatch between training and inference. Extensive experiments performed on highly expressive speech demonstrate the superiority of the proposed method in low-resource VC.

eess.AS↗