Searcharxiv⌕ Search

arXiv subjects

Yi Ren

Publications and source records attributed to Yi Ren.

At least 145 records · Page 8Linked to original sources
arXiv2023

Diffusion Denoising Process for Perceptron Bias in Out-of-distribution Detection

Out-of-distribution (OOD) detection is a crucial task for ensuring the reliability and safety of deep learning. Currently, discriminator models outperform other methods in this regard. However, the feature extraction process used by discriminator models suffers from the loss of critical information, leaving room for bad cases and malicious attacks. In this paper, we introduce a new perceptron bias assumption that suggests discriminator models are more sensitive to certain features of the input, leading to the overconfidence problem. To address this issue, we propose a novel framework that combines discriminator and generation models and integrates diffusion models (DMs) into OOD detection. We demonstrate that the diffusion denoising process (DDP) of DMs serves as a novel form of asymmetric interpolation, which is well-suited to enhance the input and mitigate the overconfidence problem. The discriminator model features of OOD data exhibit sharp changes under DDP, and we utilize the norm of this change as the indicator score. Our experiments on CIFAR10, CIFAR100, and ImageNet show that our method outperforms SOTA approaches. Notably, for the challenging InD ImageNet and OOD species datasets, our method achieves an AUROC of 85.7, surpassing the previous SOTA method's score of 77.4. Our implementation is available at \url{https://github.com/luping-liu/DiffOOD}.

cs.CV↗
arXiv2023

Detector Guidance for Multi-Object Text-to-Image Generation

Diffusion models have demonstrated impressive performance in text-to-image generation. They utilize a text encoder and cross-attention blocks to infuse textual information into images at a pixel level. However, their capability to generate images with text containing multiple objects is still restricted. Previous works identify the problem of information mixing in the CLIP text encoder and introduce the T5 text encoder or incorporate strong prior knowledge to assist with the alignment. We find that mixing problems also occur on the image side and in the cross-attention blocks. The noisy images can cause different objects to appear similar, and the cross-attention blocks inject information at a pixel level, leading to leakage of global object understanding and resulting in object mixing. In this paper, we introduce Detector Guidance (DG), which integrates a latent object detection model to separate different objects during the generation process. DG first performs latent object detection on cross-attention maps (CAMs) to obtain object information. Based on this information, DG then masks conflicting prompts and enhances related prompts by manipulating the following CAMs. We evaluate the effectiveness of DG using Stable Diffusion on COCO, CC, and a novel multi-related object benchmark, MRO. Human evaluations demonstrate that DG provides an 8-22\% advantage in preventing the amalgamation of conflicting concepts and ensuring that each object possesses its unique region without any human involvement and additional iterations. Our implementation is available at \url{https://github.com/luping-liu/Detector-Guidance}.

cs.CV↗
arXiv2023

Make-An-Audio 2: Temporal-Enhanced Text-to-Audio Generation

Large diffusion models have been successful in text-to-audio (T2A) synthesis tasks, but they often suffer from common issues such as semantic misalignment and poor temporal consistency due to limited natural language understanding and data scarcity. Additionally, 2D spatial structures widely used in T2A works lead to unsatisfactory audio quality when generating variable-length audio samples since they do not adequately prioritize temporal information. To address these challenges, we propose Make-an-Audio 2, a latent diffusion-based T2A method that builds on the success of Make-an-Audio. Our approach includes several techniques to improve semantic alignment and temporal consistency: Firstly, we use pre-trained large language models (LLMs) to parse the text into structured pairs for better temporal information capture. We also introduce another structured-text encoder to aid in learning semantic alignment during the diffusion denoising process. To improve the performance of variable length generation and enhance the temporal information extraction, we design a feed-forward Transformer-based diffusion denoiser. Finally, we use LLMs to augment and transform a large amount of audio-label data into audio-text datasets to alleviate the problem of scarcity of temporal data. Extensive experiments show that our method outperforms baseline models in both objective and subjective metrics, and achieves significant gains in temporal information understanding, semantic consistency, and sound quality.

cs.SD↗
arXiv2023

Attributing Image Generative Models using Latent Fingerprints

Generative models have enabled the creation of contents that are indistinguishable from those taken from nature. Open-source development of such models raised concerns about the risks of their misuse for malicious purposes. One potential risk mitigation strategy is to attribute generative models via fingerprinting. Current fingerprinting methods exhibit a significant tradeoff between robust attribution accuracy and generation quality while lacking design principles to improve this tradeoff. This paper investigates the use of latent semantic dimensions as fingerprints, from where we can analyze the effects of design variables, including the choice of fingerprinting dimensions, strength, and capacity, on the accuracy-quality tradeoff. Compared with previous SOTA, our method requires minimum computation and is more applicable to large-scale models. We use StyleGAN2 and the latent diffusion model to demonstrate the efficacy of our method.

cs.CV↗
arXiv2023

AV-TranSpeech: Audio-Visual Robust Speech-to-Speech Translation

Direct speech-to-speech translation (S2ST) aims to convert speech from one language into another, and has demonstrated significant progress to date. Despite the recent success, current S2ST models still suffer from distinct degradation in noisy environments and fail to translate visual speech (i.e., the movement of lips and teeth). In this work, we present AV-TranSpeech, the first audio-visual speech-to-speech (AV-S2ST) translation model without relying on intermediate text. AV-TranSpeech complements the audio stream with visual information to promote system robustness and opens up a host of practical applications: dictation or dubbing archival films. To mitigate the data scarcity with limited parallel AV-S2ST data, we 1) explore self-supervised pre-training with unlabeled audio-visual data to learn contextual representation, and 2) introduce cross-modal distillation with S2ST models trained on the audio-only corpus to further reduce the requirements of visual data. Experimental results on two language pairs demonstrate that AV-TranSpeech outperforms audio-only models under all settings regardless of the type of noise. With low-resource audio-visual data (10h, 30h), cross-modal distillation yields an improvement of 7.6 BLEU on average compared with baselines. Audio samples are available at https://AV-TranSpeech.github.io

cs.CL↗
arXiv2023

FluentSpeech: Stutter-Oriented Automatic Speech Editing with Context-Aware Diffusion Models

Stutter removal is an essential scenario in the field of speech editing. However, when the speech recording contains stutters, the existing text-based speech editing approaches still suffer from: 1) the over-smoothing problem in the edited speech; 2) lack of robustness due to the noise introduced by stutter; 3) to remove the stutters, users are required to determine the edited region manually. To tackle the challenges in stutter removal, we propose FluentSpeech, a stutter-oriented automatic speech editing model. Specifically, 1) we propose a context-aware diffusion model that iteratively refines the modified mel-spectrogram with the guidance of context features; 2) we introduce a stutter predictor module to inject the stutter information into the hidden sequence; 3) we also propose a stutter-oriented automatic speech editing (SASE) dataset that contains spontaneous speech recordings with time-aligned stutter labels to train the automatic stutter localization model. Experimental results on VCTK and LibriTTS datasets demonstrate that our model achieves state-of-the-art performance on speech editing. Further experiments on our SASE dataset show that FluentSpeech can effectively improve the fluency of stuttering speech in terms of objective and subjective metrics. Code and audio samples can be found at https://github.com/Zain-Jiang/Speech-Editing-Toolkit.

cs.SD↗
arXiv2023

FedSpeech: Federated Text-to-Speech with Continual Learning

Federated learning enables collaborative training of machine learning models under strict privacy restrictions and federated text-to-speech aims to synthesize natural speech of multiple users with a few audio training samples stored in their devices locally. However, federated text-to-speech faces several challenges: very few training samples from each speaker are available, training samples are all stored in local device of each user, and global model is vulnerable to various attacks. In this paper, we propose a novel federated learning architecture based on continual learning approaches to overcome the difficulties above. Specifically, 1) we use gradual pruning masks to isolate parameters for preserving speakers' tones; 2) we apply selective masks for effectively reusing knowledge from tasks; 3) a private speaker embedding is introduced to keep users' privacy. Experiments on a reduced VCTK dataset demonstrate the effectiveness of FedSpeech: it nearly matches multi-task training in terms of multi-speaker speech quality; moreover, it sufficiently retains the speakers' tones and even outperforms the multi-task training in the speaker similarity experiment.

eess.AS↗
arXiv2023

CLAPSpeech: Learning Prosody from Text Context with Contrastive Language-Audio Pre-training

Improving text representation has attracted much attention to achieve expressive text-to-speech (TTS). However, existing works only implicitly learn the prosody with masked token reconstruction tasks, which leads to low training efficiency and difficulty in prosody modeling. We propose CLAPSpeech, a cross-modal contrastive pre-training framework that explicitly learns the prosody variance of the same text token under different contexts. Specifically, 1) We encourage the model to connect the text context with its corresponding prosody pattern in the joint multi-modal space with the elaborate design of the encoder inputs and contrastive loss; 2) We introduce a multi-scale pre-training pipeline to capture prosody patterns in multiple levels. We show how to incorporate CLAPSpeech into existing TTS models for better prosody. Experiments on three datasets not only show that CLAPSpeech could improve the prosody prediction for existing TTS methods, but also demonstrate its generalization ability to adapt to multiple languages and multi-speaker TTS. We also deeply analyze the principle behind the performance of CLAPSpeech. Ablation studies demonstrate the necessity of each component in our method. Source code and audio samples are available at https://clapspeech.github.io.

cs.SD↗
arXiv2023

GeneFace++: Generalized and Stable Real-Time Audio-Driven 3D Talking Face Generation

Generating talking person portraits with arbitrary speech audio is a crucial problem in the field of digital human and metaverse. A modern talking face generation method is expected to achieve the goals of generalized audio-lip synchronization, good video quality, and high system efficiency. Recently, neural radiance field (NeRF) has become a popular rendering technique in this field since it could achieve high-fidelity and 3D-consistent talking face generation with a few-minute-long training video. However, there still exist several challenges for NeRF-based methods: 1) as for the lip synchronization, it is hard to generate a long facial motion sequence of high temporal consistency and audio-lip accuracy; 2) as for the video quality, due to the limited data used to train the renderer, it is vulnerable to out-of-domain input condition and produce bad rendering results occasionally; 3) as for the system efficiency, the slow training and inference speed of the vanilla NeRF severely obstruct its usage in real-world applications. In this paper, we propose GeneFace++ to handle these challenges by 1) utilizing the pitch contour as an auxiliary feature and introducing a temporal loss in the facial motion prediction process; 2) proposing a landmark locally linear embedding method to regulate the outliers in the predicted motion sequence to avoid robustness issues; 3) designing a computationally efficient NeRF-based motion-to-video renderer to achieves fast training and real-time inference. With these settings, GeneFace++ becomes the first NeRF-based method that achieves stable and real-time talking face generation with generalized audio-lip synchronization. Extensive experiments show that our method outperforms state-of-the-art baselines in terms of subjective and objective evaluation. Video samples are available at https://genefaceplusplus.github.io .

cs.CV↗
arXiv2023

AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking Head

Large language models (LLMs) have exhibited remarkable capabilities across a variety of domains and tasks, challenging our understanding of learning and cognition. Despite the recent success, current LLMs are not capable of processing complex audio information or conducting spoken conversations (like Siri or Alexa). In this work, we propose a multi-modal AI system named AudioGPT, which complements LLMs (i.e., ChatGPT) with 1) foundation models to process complex audio information and solve numerous understanding and generation tasks; and 2) the input/output interface (ASR, TTS) to support spoken dialogue. With an increasing demand to evaluate multi-modal LLMs of human intention understanding and cooperation with foundation models, we outline the principles and processes and test AudioGPT in terms of consistency, capability, and robustness. Experimental results demonstrate the capabilities of AudioGPT in solving AI tasks with speech, music, sound, and talking head understanding and generation in multi-round dialogues, which empower humans to create rich and diverse audio content with unprecedented ease. Our system is publicly available at \url{https://github.com/AIGC-Audio/AudioGPT}.

cs.CL↗
arXiv2023

Unbiased Pairwise Learning from Implicit Feedback for Recommender Systems without Biased Variance Control

Generally speaking, the model training for recommender systems can be based on two types of data, namely explicit feedback and implicit feedback. Moreover, because of its general availability, we see wide adoption of implicit feedback data, such as click signal. There are mainly two challenges for the application of implicit feedback. First, implicit data just includes positive feedback. Therefore, we are not sure whether the non-interacted items are really negative or positive but not displayed to the corresponding user. Moreover, the relevance of rare items is usually underestimated since much fewer positive feedback of rare items is collected compared with popular ones. To tackle such difficulties, both pointwise and pairwise solutions are proposed before for unbiased relevance learning. As pairwise learning suits well for the ranking tasks, the previously proposed unbiased pairwise learning algorithm already achieves state-of-the-art performance. Nonetheless, the existing unbiased pairwise learning method suffers from high variance. To get satisfactory performance, non-negative estimator is utilized for practical variance control but introduces additional bias. In this work, we propose an unbiased pairwise learning method, named UPL, with much lower variance to learn a truly unbiased recommender model. Extensive offline experiments on real world datasets and online A/B testing demonstrate the superior performance of our proposed method.

cs.IR↗
arXiv2023

Evolved Massive Stars at Low-metallicity V. Mass-Loss Rate of Red Supergiant Stars in the Small Magellanic Cloud

We assemble the most complete and clean red supergiant (RSG) sample (2,121 targets) so far in the Small Magellanic Cloud (SMC) with 53 different bands of data to study the MLR of RSGs. In order to match the observed spectral energy distributions (SEDs), a theoretical grid of 17,820 Oxygen-rich models (``normal'' and ``dusty'' grids are half-and-half) is created by the radiatively-driven wind model of the DUSTY code, covering a wide range of dust parameters. We select the best model for each target by calculating the minimal modified chi-square and visual inspection. The resulting MLRs from DUSTY are converted to real MLRs based on the scaling relation, for which a total MLR of $6.16\times10^{-3}$ $M_\odot$ yr$^{-1}$ is measured (corresponding to a dust-production rate of $\sim6\times10^{-6}$ $M_\odot$ yr$^{-1}$), with a typical MLR of $\sim10^{-6}$ $M_\odot$ yr$^{-1}$ for the general population of the RSGs. The complexity of mass-loss estimation based on the SED is fully discussed for the first time, indicating large uncertainties based on the photometric data (potentially up to one order of magnitude or more). The Hertzsprung-Russell and luminosity versus median absolute deviation diagrams of the sample indicate the positive relation between luminosity and MLR. Meanwhile, the luminosity versus MLR diagrams show a ``knee-like'' shape with enhanced mass-loss occurring above $\log_{10}(L/L_\odot)\approx4.6$, which may be due to the degeneracy of luminosity, pulsation, low surface gravity, convection, and other factors. We derive our MLR relation by using a third-order polynomial to fit the sample and compare our result with previous empirical MLR prescriptions. Given that our MLR prescription is based on a much larger sample than previous determinations, it provides a more accurate relation at the cool and luminous region of the H-R diagram at low-metallicity compared to previous studies.

astro-ph.SR↗
arXiv2023

MUG: A General Meeting Understanding and Generation Benchmark

Listening to long video/audio recordings from video conferencing and online courses for acquiring information is extremely inefficient. Even after ASR systems transcribe recordings into long-form spoken language documents, reading ASR transcripts only partly speeds up seeking information. It has been observed that a range of NLP applications, such as keyphrase extraction, topic segmentation, and summarization, significantly improve users' efficiency in grasping important information. The meeting scenario is among the most valuable scenarios for deploying these spoken language processing (SLP) capabilities. However, the lack of large-scale public meeting datasets annotated for these SLP tasks severely hinders their advancement. To prompt SLP advancement, we establish a large-scale general Meeting Understanding and Generation Benchmark (MUG) to benchmark the performance of a wide range of SLP tasks, including topic segmentation, topic-level and session-level extractive summarization and topic title generation, keyphrase extraction, and action item detection. To facilitate the MUG benchmark, we construct and release a large-scale meeting dataset for comprehensive long-form SLP development, the AliMeeting4MUG Corpus, which consists of 654 recorded Mandarin meeting sessions with diverse topic coverage, with manual annotations for SLP tasks on manual transcripts of meeting recordings. To the best of our knowledge, the AliMeeting4MUG Corpus is so far the largest meeting corpus in scale and facilitates most SLP tasks. In this paper, we provide a detailed introduction of this corpus, SLP tasks and evaluation methods, baseline systems and their performance.

cs.CL↗
arXiv2023

Overview of the ICASSP 2023 General Meeting Understanding and Generation Challenge (MUG)

ICASSP2023 General Meeting Understanding and Generation Challenge (MUG) focuses on prompting a wide range of spoken language processing (SLP) research on meeting transcripts, as SLP applications are critical to improve users' efficiency in grasping important information in meetings. MUG includes five tracks, including topic segmentation, topic-level and session-level extractive summarization, topic title generation, keyphrase extraction, and action item detection. To facilitate MUG, we construct and release a large-scale meeting dataset, the AliMeeting4MUG Corpus.

cs.CL↗
arXiv2023

Unbiased Learning to Rank with Biased Continuous Feedback

It is a well-known challenge to learn an unbiased ranker with biased feedback. Unbiased learning-to-rank(LTR) algorithms, which are verified to model the relative relevance accurately based on noisy feedback, are appealing candidates and have already been applied in many applications with single categorical labels, such as user click signals. Nevertheless, the existing unbiased LTR methods cannot properly handle continuous feedback, which are essential for many industrial applications, such as content recommender systems. To provide personalized high-quality recommendation results, recommender systems need model both categorical and continuous biased feedback, such as click and dwell time. Accordingly, we design a novel unbiased LTR algorithm to tackle the challenges, which innovatively models position bias in the pairwise fashion and introduces the pairwise trust bias to separate the position bias, trust bias, and user relevance explicitly and can work for both continuous and categorical feedback. Experiment results on public benchmark datasets and internal live traffic of a large-scale recommender system at Tencent News show superior results for continuous labels and also competitive performance for categorical labels of the proposed method.

cs.IR↗
arXiv2023

An Improved Pair Method to Probe the Dust Extinction Law

Dust extinction law is crucial to recover the intrinsic energy distribution of celestial objects and infer the characteristics of interstellar dust. Based on the traditional pair method, an improved pair method is proposed to model the dust extinguished spectral energy distribution (SED) of an individual star. Instead of the mathematically parameterizing extinction curves, the extinction curves in this work are directly from the silicate-graphite dust model, so that the dust extinction law can be obtained and the dust properties can be analyzed simultaneously. The ATLAS9 stellar model atmosphere is adopted for the intrinsic SEDs in this work, while the silicate-graphite dust model with a dust size distribution of $dn/da \sim a^{-α}{\rm exp}(-a/a_c),~0.005 < a < 5~μ{\rm m}$ for each component is adopted for the model extinction curves. One typical extinction tracer in the dense region (V410 Anon9) and one in the diffuse region (Cyg OB2 \#12) of the MW are chosen to test the reliability and the practicability of the improved pair method in different stellar environments. The results are consistent with their interstellar environments and are in agreement with the previous observations and studies, which prove that the improved pair method is effective and applicable in different stellar environments. In addition to the reliable extinction results, the derived parameters in the dust model can be used to analyze the dust properties, which cannot be achieved by other methods with the mathematical extinction models. With the improved pair method, the stellar parameters can also be inferred and the extinction law beyond the wavelengths of observed data can be predicted based on the dust model as well.

astro-ph.IM↗
arXiv2023

TranSpeech: Speech-to-Speech Translation With Bilateral Perturbation

Direct speech-to-speech translation (S2ST) with discrete units leverages recent progress in speech representation learning. Specifically, a sequence of discrete representations derived in a self-supervised manner are predicted from the model and passed to a vocoder for speech reconstruction, while still facing the following challenges: 1) Acoustic multimodality: the discrete units derived from speech with same content could be indeterministic due to the acoustic property (e.g., rhythm, pitch, and energy), which causes deterioration of translation accuracy; 2) high latency: current S2ST systems utilize autoregressive models which predict each unit conditioned on the sequence previously generated, failing to take full advantage of parallelism. In this work, we propose TranSpeech, a speech-to-speech translation model with bilateral perturbation. To alleviate the acoustic multimodal problem, we propose bilateral perturbation (BiP), which consists of the style normalization and information enhancement stages, to learn only the linguistic information from speech samples and generate more deterministic representations. With reduced multimodality, we step forward and become the first to establish a non-autoregressive S2ST technique, which repeatedly masks and predicts unit choices and produces high-accuracy results in just a few cycles. Experimental results on three language pairs demonstrate that BiP yields an improvement of 2.9 BLEU on average compared with a baseline textless S2ST model. Moreover, our parallel decoding shows a significant reduction of inference latency, enabling speedup up to 21.4x than autoregressive technique. Audio samples are available at \url{https://TranSpeech.github.io/}

cs.CL↗
arXiv2023

Toward Global Sensing Quality Maximization: A Configuration Optimization Scheme for Camera Networks

The performance of a camera network monitoring a set of targets depends crucially on the configuration of the cameras. In this paper, we investigate the reconfiguration strategy for the parameterized camera network model, with which the sensing qualities of the multiple targets can be optimized globally and simultaneously. We first propose to use the number of pixels occupied by a unit-length object in image as a metric of the sensing quality of the object, which is determined by the parameters of the camera, such as intrinsic, extrinsic, and distortional coefficients. Then, we form a single quantity that measures the sensing quality of the targets by the camera network. This quantity further serves as the objective function of our optimization problem to obtain the optimal camera configuration. We verify the effectiveness of our approach through extensive simulations and experiments, and the results reveal its improved performance on the AprilTag detection tasks. Codes and related utilities for this work are open-sourced and available at https://github.com/sszxc/MultiCam-Simulation.

cs.CV↗