Anurag Kumar

Muqiao Yang, Joseph Konan, David Bick et al. · cmu

While deep learning based speech enhancement systems have made rapid progress in improving the quality of speech signals, they can still produce outputs that contain artifacts and can sound unnatural. We propose a novel approach to speech enhancement aimed at improving perceptual quality and naturalness of enhanced signals by optimizing for key characteristics of speech. We first identify key acoustic parameters that have been found to correlate well with voice quality (e.g. jitter, shimmer, and spectral flux) and then propose objective functions which are aimed at reducing the difference between clean speech and enhanced speech with respect to these features. The full set of acoustic features is the extended Geneva Acoustic Parameter Set (eGeMAPS), which includes 25 different attributes associated with perception of speech. Given the non-differentiable nature of these feature computation, we first build differentiable estimators of the eGeMAPS and then use them to fine-tune existing speech enhancement systems. Our approach is generic and can be applied to any existing deep learning based enhancement systems to further improve the enhanced speech signals. Experimental results conducted on the Deep Noise Suppression (DNS) Challenge dataset shows that our approach can improve the state-of-the-art deep learning based enhancement systems.

Yunyang Zeng, Joseph Konan, Shuo Han et al. · cmu

Speech enhancement models have greatly progressed in recent years, but still show limits in perceptual quality of their speech outputs. We propose an objective for perceptual quality based on temporal acoustic parameters. These are fundamental speech features that play an essential role in various applications, including speaker recognition and paralinguistic analysis. We provide a differentiable estimator for four categories of low-level acoustic descriptors involving: frequency-related parameters, energy or amplitude-related parameters, spectral balance parameters, and temporal features. Unlike prior work that looks at aggregated acoustic parameters or a few categories of acoustic parameters, our temporal acoustic parameter (TAP) loss enables auxiliary optimization and improvement of many fine-grain speech characteristics in enhancement workflows. We show that adding TAPLoss as an auxiliary objective in speech enhancement produces speech with improved perceptual quality and intelligibility. We use data from the Deep Noise Suppression 2020 Challenge to demonstrate that both time-domain models and time-frequency domain models can benefit from our method.

Muqiao Yang, Joseph Konan, David Bick et al. · cmu

Despite rapid advancement in recent years, current speech enhancement models often produce speech that differs in perceptual quality from real clean speech. We propose a learning objective that formalizes differences in perceptual quality, by using domain knowledge of acoustic-phonetics. We identify temporal acoustic parameters -- such as spectral tilt, spectral flux, shimmer, etc. -- that are non-differentiable, and we develop a neural network estimator that can accurately predict their time-series values across an utterance. We also model phoneme-specific weights for each feature, as the acoustic parameters are known to show different behavior in different phonemes. We can add this criterion as an auxiliary loss to any model that produces speech, to optimize speech outputs to match the values of clean speech in these features. Experimentally we show that it improves speech enhancement workflows in both time-domain and time-frequency domain, as measured by standard evaluation metrics. We also provide an analysis of phoneme-dependent improvement on acoustic parameters, demonstrating the additional interpretability that our method provides. This analysis can suggest which features are currently the bottleneck for improvement.

Rodrigo Mira, Buye Xu, Jacob Donley et al.

Audio-visual speech enhancement aims to extract clean speech from a noisy environment by leveraging not only the audio itself but also the target speaker's lip movements. This approach has been shown to yield improvements over audio-only speech enhancement, particularly for the removal of interfering speech. Despite recent advances in speech synthesis, most audio-visual approaches continue to use spectral mapping/masking to reproduce the clean audio, often resulting in visual backbones added to existing speech enhancement architectures. In this work, we propose LA-VocE, a new two-stage approach that predicts mel-spectrograms from noisy audio-visual speech via a transformer-based architecture, and then converts them into waveform audio using a neural vocoder (HiFi-GAN). We train and evaluate our framework on thousands of speakers and 11+ different languages, and study our model's ability to adapt to different levels of background noise and speech interference. Our experiments show that LA-VocE outperforms existing methods according to multiple metrics, particularly under very noisy scenarios.

Chandramani Singh, Eitan Altman, Anurag Kumar et al.

We study the trade-off between delivery delay and energy consumption in a delay tolerant network in which a message (or a file) has to be delivered to each of several destinations by epidemic relaying. In addition to the destinations, there are several other nodes in the network that can assist in relaying the message. We first assume that, at every instant, all the nodes know the number of relays carrying the packet and the number of destinations that have received the packet. We formulate the problem as a controlled continuous time Markov chain and derive the optimal closed loop control (i.e., forwarding policy). However, in practice, the intermittent connectivity in the network implies that the nodes may not have the required perfect knowledge of the system state. To address this issue, we obtain an ODE (i.e., a deterministic fluid) approximation for the optimally controlled Markov chain. This fluid approximation also yields an asymptotically optimal open loop policy. Finally, we evaluate the performance of the deterministic policy over finite networks. Numerical results show that this policy performs close to the optimal closed loop policy.

Susan Liang, Chao Huang, Yapeng Tian et al.

Can machines recording an audio-visual scene produce realistic, matching audio-visual experiences at novel positions and novel view directions? We answer it by studying a new task -- real-world audio-visual scene synthesis -- and a first-of-its-kind NeRF-based approach for multimodal learning. Concretely, given a video recording of an audio-visual scene, the task is to synthesize new videos with spatial audios along arbitrary novel camera trajectories in that scene. We propose an acoustic-aware audio generation module that integrates prior knowledge of audio propagation into NeRF, in which we implicitly associate audio generation with the 3D geometry and material properties of a visual environment. Furthermore, we present a coordinate transformation module that expresses a view direction relative to the sound source, enabling the model to learn sound source-centric acoustic fields. To facilitate the study of this new task, we collect a high-quality Real-World Audio-Visual Scene (RWAVS) dataset. We demonstrate the advantages of our method on this real-world dataset and the simulation-based SoundSpaces dataset.

Chao Huang, Yapeng Tian, Anurag Kumar et al.

Humans naturally perceive surrounding scenes by unifying sound and sight in a first-person view. Likewise, machines are advanced to approach human intelligence by learning with multisensory inputs from an egocentric perspective. In this paper, we explore the challenging egocentric audio-visual object localization task and observe that 1) egomotion commonly exists in first-person recordings, even within a short duration; 2) The out-of-view sound components can be created while wearers shift their attention. To address the first problem, we propose a geometry-aware temporal aggregation module to handle the egomotion explicitly. The effect of egomotion is mitigated by estimating the temporal geometry transformation and exploiting it to update visual representations. Moreover, we propose a cascaded feature enhancement module to tackle the second issue. It improves cross-modal localization robustness by disentangling visually-indicated audio representation. During training, we take advantage of the naturally available audio-visual temporal synchronization as the ``free'' self-supervision to avoid costly labeling. We also annotate and create the Epic Sounding Object dataset for evaluation purposes. Extensive experiments show that our method achieves state-of-the-art localization performance in egocentric videos and can be generalized to diverse audio-visual scenes.

Susan Liang, Chao Huang, Yapeng Tian et al.

Room impulse response (RIR), which measures the sound propagation within an environment, is critical for synthesizing high-fidelity audio for a given environment. Some prior work has proposed representing RIR as a neural field function of the sound emitter and receiver positions. However, these methods do not sufficiently consider the acoustic properties of an audio scene, leading to unsatisfactory performance. This letter proposes a novel Neural Acoustic Context Field approach, called NACF, to parameterize an audio scene by leveraging multiple acoustic contexts, such as geometry, material property, and spatial information. Driven by the unique properties of RIR, i.e., temporal un-smoothness and monotonic energy attenuation, we design a temporal correlation module and multi-scale energy decay criterion. Experimental results show that NACF outperforms existing field-based methods by a notable margin. Please visit our project page for more qualitative results.

Chao Huang, Susan Liang, Yapeng Tian et al.

We propose DAVIS, a Diffusion-based Audio-VIsual Separation framework that solves the audio-visual sound source separation task through generative learning. Existing methods typically frame sound separation as a mask-based regression problem, achieving significant progress. However, they face limitations in capturing the complex data distribution required for high-quality separation of sounds from diverse categories. In contrast, DAVIS leverages a generative diffusion model and a Separation U-Net to synthesize separated sounds directly from Gaussian noise, conditioned on both the audio mixture and the visual information. With its generative objective, DAVIS is better suited to achieving the goal of high-quality sound separation across diverse sound categories. We compare DAVIS to existing state-of-the-art discriminative audio-visual separation methods on the AVE and MUSIC datasets, and results show that DAVIS outperforms other methods in separation quality, demonstrating the advantages of our framework for tackling the audio-visual source separation task.

Haibin Wu, Ke Tan, Buye Xu et al.

Despite multiple efforts made towards adopting complex-valued deep neural networks (DNNs), it remains an open question whether complex-valued DNNs are generally more effective than real-valued DNNs for monaural speech enhancement. This work is devoted to presenting a critical assessment by systematically examining complex-valued DNNs against their real-valued counterparts. Specifically, we investigate complex-valued DNN atomic units, including linear layers, convolutional layers, long short-term memory (LSTM), and gated linear units. By comparing complex- and real-valued versions of fundamental building blocks in the recently developed gated convolutional recurrent network (GCRN), we show how different mechanisms for basic blocks affect the performance. We also find that the use of complex-valued operations hinders the model capacity when the model size is small. In addition, we examine two recent complex-valued DNNs, i.e. deep complex convolutional recurrent network (DCCRN) and deep complex U-Net (DCUNET). Evaluation results show that both DNNs produce identical performance to their real-valued counterparts while requiring much more computation. Based on these comprehensive comparisons, we conclude that complex-valued DNNs do not provide a performance gain over their real-valued counterparts for monaural speech enhancement, and thus are less desirable due to their higher computational costs.

Heeseung Yun, Ruohan Gao, Ishwarya Ananthabhotla et al.

Egocentric videos provide comprehensive contexts for user and scene understanding, spanning multisensory perception to behavioral interaction. We propose Spherical World-Locking (SWL) as a general framework for egocentric scene representation, which implicitly transforms multisensory streams with respect to measurements of head orientation. Compared to conventional head-locked egocentric representations with a 2D planar field-of-view, SWL effectively offsets challenges posed by self-motion, allowing for improved spatial synchronization between input modalities. Using a set of multisensory embeddings on a worldlocked sphere, we design a unified encoder-decoder transformer architecture that preserves the spherical structure of the scene representation, without requiring expensive projections between image and world coordinate systems. We evaluate the effectiveness of the proposed framework on multiple benchmark tasks for egocentric video understanding, including audio-visual active speaker localization, auditory spherical source localization, and behavior anticipation in everyday activities.

Kuan-Lin Chen, Daniel D. E. Wong, Ke Tan et al.

Most speech enhancement (SE) models learn a point estimate and do not make use of uncertainty estimation in the learning process. In this paper, we show that modeling heteroscedastic uncertainty by minimizing a multivariate Gaussian negative log-likelihood (NLL) improves SE performance at no extra cost. During training, our approach augments a model learning complex spectral mapping with a temporary submodel to predict the covariance of the enhancement error at each time-frequency bin. Due to unrestricted heteroscedastic uncertainty, the covariance introduces an undersampling effect, detrimental to SE performance. To mitigate undersampling, our approach inflates the uncertainty lower bound and weights each loss component with their uncertainty, effectively compensating severely undersampled components with more penalties. Our multivariate setting reveals common covariance assumptions such as scalar and diagonal matrices. By weakening these assumptions, we show that the NLL achieves superior performance compared to popular loss functions including the mean squared error (MSE), mean absolute error (MAE), and scale-invariant signal-to-distortion ratio (SI-SDR).

Srinivasan Venkatramanan, Anurag Kumar

We consider a setting in which a single item of content (such as a song or a video clip) is disseminated in a population of mobile nodes by opportunistic copying when pairs of nodes come in radio contact. We propose and study models that capture the joint evolution of the population of nodes interested in the content (referred to as destinations), and the population of nodes that possess the content. The evolution of interest in the content is captured using an influence spread model and the content spread occurs via epidemic copying. Nodes not yet interested in the content are called relays; the influence spread process converts relays into destinations. We consider the decentralized setting, where interest in the content and the spread of the content evolve by pairwise interactions between the mobiles. We derive fluid limits for the joint evolution models and obtain optimal policies for copying to relay nodes in order to deliver content to a desired fraction of destinations. We prove that a time-threshold policy is optimal while copying to relays. We then provide insights into the effects of various system parameters on the co-evolution model through simulations.

Anurag Kumar, Raghuveer Peri, Jon Burnsky et al.

Multimodal large-language models (MLLMs) often experience degraded safety alignment when harmful queries exploit cross-modal interactions. Models aligned on text alone show a higher rate of successful attacks when extended to two or more modalities. In this work, we propose a simple conditional decoding strategy, CASA (Classification Augmented with Safety Attention) that utilizes internal representations of MLLMs to predict a binary safety token before response generation. We introduce a novel safety attention module designed to enhance the model's ability to detect malicious queries. Our design ensures robust safety alignment without relying on any external classifier or auxiliary head, and without the need for modality-specific safety fine-tuning. On diverse benchmarks such as MM-SafetyBench, JailbreakV-28k, and adversarial audio tests, CASA lowers the average attack success rate by more than 97% across modalities and across attack types. Our empirical evaluations also show that CASA maintains strong utility in benign inputs, a result validated through both automated and human evaluations (via 13 trained annotators). Together, these results highlight CASA as a simple and generalizable framework to improve multimodal LLM safety.

Gheorghe Comanici, Eric Bieber, Mike Schaekermann et al. · amazon-science, baidu

In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving.

Ziyang Chen, Israel D. Gebru, Christian Richardt et al.

We present a new dataset called Real Acoustic Fields (RAF) that captures real acoustic room data from multiple modalities. The dataset includes high-quality and densely captured room impulse response data paired with multi-view images, and precise 6DoF pose tracking data for sound emitters and listeners in the rooms. We used this dataset to evaluate existing methods for novel-view acoustic synthesis and impulse response generation which previously relied on synthetic data. In our evaluation, we thoroughly assessed existing audio and audio-visual models against multiple criteria and proposed settings to enhance their performance on real-world data. We also conducted experiments to investigate the impact of incorporating visual data (i.e., images and depth) into neural acoustic field models. Additionally, we demonstrated the effectiveness of a simple sim2real approach, where a model is pre-trained with simulated data and fine-tuned with sparse real-world data, resulting in significant improvements in the few-shot learning approach. RAF is the first dataset to provide densely captured room acoustic data, making it an ideal resource for researchers working on audio and audio-visual neural acoustic field modeling techniques. Demos and datasets are available on our project page: https://facebookresearch.github.io/real-acoustic-fields/

Chao Huang, Susan Liang, Yunlong Tang et al.

Text-guided diffusion models have revolutionized generative tasks by producing high-fidelity content from text descriptions. They have also enabled an editing paradigm where concepts can be replaced through text conditioning (e.g., a dog to a tiger). In this work, we explore a novel approach: instead of replacing a concept, can we enhance or suppress the concept itself? Through an empirical study, we identify a trend where concepts can be decomposed in text-guided diffusion models. Leveraging this insight, we introduce ScalingConcept, a simple yet effective method to scale decomposed concepts up or down in real input without introducing new elements. To systematically evaluate our approach, we present the WeakConcept-10 dataset, where concepts are imperfect and need to be enhanced. More importantly, ScalingConcept enables a variety of novel zero-shot applications across image and audio domains, including tasks such as canonical pose generation and generative sound highlighting or removal.

Xiulong Liu, Anurag Kumar, Paul Calamia et al.

In mixed reality applications, a realistic acoustic experience in spatial environments is as crucial as the visual experience for achieving true immersion. Despite recent advances in neural approaches for Room Impulse Response (RIR) estimation, most existing methods are limited to the single environment on which they are trained, lacking the ability to generalize to new rooms with different geometries and surface materials. We aim to develop a unified model capable of reconstructing the spatial acoustic experience of any environment with minimum additional measurements. To this end, we present xRIR, a framework for cross-room RIR prediction. The core of our generalizable approach lies in combining a geometric feature extractor, which captures spatial context from panorama depth images, with a RIR encoder that extracts detailed acoustic features from only a few reference RIR samples. To evaluate our method, we introduce ACOUSTICROOMS, a new dataset featuring high-fidelity simulation of over 300,000 RIRs from 260 rooms. Experiments show that our method strongly outperforms a series of baselines. Furthermore, we successfully perform sim-to-real transfer by evaluating our model on four real-world environments, demonstrating the generalizability of our approach and the realism of our dataset.

Ravi Shankar, Ke Tan, Buye Xu et al.

Self-supervised learned models have been found to be very effective for certain speech tasks such as automatic speech recognition, speaker identification, keyword spotting and others. While the features are undeniably useful in speech recognition and associated tasks, their utility in speech enhancement systems is yet to be firmly established, and perhaps not properly understood. In this paper, we investigate the uses of SSL representations for single-channel speech enhancement in challenging conditions and find that they add very little value for the enhancement task. Our constraints are designed around on-device real-time speech enhancement -- model is causal, the compute footprint is small. Additionally, we focus on low SNR conditions where such models struggle to provide good enhancement. In order to systematically examine how SSL representations impact performance of such enhancement models, we propose a variety of techniques to utilize these embeddings which include different forms of knowledge-distillation and pre-training.

Chao Huang, Ruohan Gao, J. M. F. Tsang et al.

Recent years have seen a significant increase in video content creation and consumption. Crafting engaging content requires the careful curation of both visual and audio elements. While visual cue curation, through techniques like optimal viewpoint selection or post-editing, has been central to media production, its natural counterpart, audio, has not undergone equivalent advancements. This often results in a disconnect between visual and acoustic saliency. To bridge this gap, we introduce a novel task: visually-guided acoustic highlighting, which aims to transform audio to deliver appropriate highlighting effects guided by the accompanying video, ultimately creating a more harmonious audio-visual experience. We propose a flexible, transformer-based multimodal framework to solve this task. To train our model, we also introduce a new dataset -- the muddy mix dataset, leveraging the meticulous audio and video crafting found in movies, which provides a form of free supervision. We develop a pseudo-data generation process to simulate poorly mixed audio, mimicking real-world scenarios through a three-step process -- separation, adjustment, and remixing. Our approach consistently outperforms several baselines in both quantitative and subjective evaluation. We also systematically study the impact of different types of contextual guidance and difficulty levels of the dataset. Our project page is here: https://wikichao.github.io/VisAH/.

Anurag Kumar, Rohit Paturi, Amber Afshan et al.

Speaker Diarization (SD) is a crucial component of modern end-to-end ASR pipelines. Traditional SD systems, which are typically audio-based and operate independently of ASR, often introduce speaker errors, particularly during speaker transitions and overlapping speech. Recently, language models including fine-tuned large language models (LLMs) have shown to be effective as a second-pass speaker error corrector by leveraging lexical context in the transcribed output. In this work, we introduce a novel acoustic conditioning approach to provide more fine-grained information from the acoustic diarizer to the LLM. We also show that a simpler constrained decoding strategy reduces LLM hallucinations, while avoiding complicated post-processing. Our approach significantly reduces the speaker error rates by 24-43% across Fisher, Callhome, and RT03-CTS datasets, compared to the first-pass Acoustic SD.

Artem Dementyev, Wazeer Zulfikar, Sinan Hersek et al.

Current multimodal LLMs process audio as a mono stream, ignoring the rich spatial information essential for embodied AI. Existing spatial audio models, conversely, are constrained to fixed microphone geometries, preventing deployment across diverse devices. We present PhaseCoder, a transformer-only spatial audio encoder that is agnostic to microphone geometry. PhaseCoder takes raw multichannel audio and microphone coordinates as inputs to perform localization and produces robust spatial embeddings. We demonstrate that Gemma 3n LLM can be fine-tuned to reason over "Spatial Audio Tokens" produced by PhaseCoder. We show our encoder achieves state-of-the-art results on microphone-invariant localization benchmarks and, for the first time, enables an LLM to perform complex spatial reasoning and targeted transcription tasks from an arbitrary microphone array.

Chao Huang, Susan Liang, Yapeng Tian et al.

We propose DAVIS, a Diffusion-based Audio-VIsual Separation framework that solves the audio-visual sound source separation task through generative learning. Existing methods typically frame sound separation as a mask-based regression problem, achieving significant progress. However, they face limitations in capturing the complex data distribution required for high-quality separation of sounds from diverse categories. In contrast, DAVIS circumvents these issues by leveraging potent generative modeling paradigms, specifically Denoising Diffusion Probabilistic Models (DDPM) and the more recent Flow Matching (FM), integrated within a specialized Separation U-Net architecture. Our framework operates by synthesizing the desired separated sound spectrograms directly from a noise distribution, conditioned concurrently on the mixed audio input and associated visual information. The inherent nature of its generative objective makes DAVIS particularly adept at producing high-quality sound separations for diverse sound categories. We present comparative evaluations of DAVIS, encompassing both its DDPM and Flow Matching variants, against leading methods on the standard AVE and MUSIC datasets. The results affirm that both variants surpass existing approaches in separation quality, highlighting the efficacy of our generative framework for tackling the audio-visual source separation task.

Joanna Hong, Sanjeel Parekh, Honglie Chen et al.

Building reliable speech systems often requires combining multiple modalities, like audio and visual cues. While such multimodal solutions frequently lead to improvements in performance and may even be critical in certain cases, they come with several constraints such as increased sensory requirements, computational cost, and modality synchronization, to mention a few. These challenges constrain the direct uses of these multimodal solutions in real-world applications. In this work, we develop approaches where the learning happens with all available modalities but the deployment or inference is done with just one or reduced modalities. To do so, we propose a Multimodal Training and Unimodal Deployment (MUTUD) framework which includes a Temporally Aligned Modality feature Estimation (TAME) module that can estimate information from missing modality using modalities present during inference. This innovative approach facilitates the integration of information across different modalities, enhancing the overall inference process by leveraging the strengths of each modality to compensate for the absence of certain modalities during inference. We apply MUTUD to various audiovisual speech tasks and show that it can reduce the performance gap between the multimodal and corresponding unimodal models to a considerable extent. MUTUD can achieve this while reducing the model size and compute compared to multimodal models, in some cases by almost 80%.

Shrestha Mohanty, Sarah Xuan, Jacob Jobraeel et al.

We focus on enhancing comprehension in small-group recorded conversations, which serve as a medium to bring people together and provide a space for sharing personal stories and experiences on crucial social matters. One way to parse and convey information from these conversations is by sharing highlighted excerpts in subsequent conversations. This can help promote a collective understanding of relevant issues, by highlighting perspectives and experiences to other groups of people who might otherwise be unfamiliar with and thus unable to relate to these experiences. The primary challenge that arises then is that excerpts taken from one conversation and shared in another setting might be missing crucial context or key elements that were previously introduced in the original conversation. This problem is exacerbated when conversations become lengthier and richer in themes and shared experiences. To address this, we explore how Large Language Models (LLMs) can enrich these excerpts by providing socially relevant context. We present approaches for effective contextualization to improve comprehension, readability, and empathy. We show significant improvements in understanding, as assessed through subjective and objective evaluations. While LLMs can offer valuable context, they struggle with capturing key social aspects. We release the Human-annotated Salient Excerpts (HSE) dataset to support future work. Additionally, we show how context-enriched excerpts can provide more focused and comprehensive conversation summaries.

Vahid Ahmadi Kalkhorani, Cheng Yu, Anurag Kumar et al.

Adding visual cues to audio-based speech separation can improve separation performance. This paper introduces AV-CrossNet, an audiovisual (AV) system for speech enhancement, target speaker extraction, and multi-talker speaker separation. AV-CrossNet is extended from the CrossNet architecture, which is a recently proposed network that performs complex spectral mapping for speech separation by leveraging global attention and positional encoding. To effectively utilize visual cues, the proposed system incorporates pre-extracted visual embeddings and employs a visual encoder comprising temporal convolutional layers. Audio and visual features are fused in an early fusion layer before feeding to AV-CrossNet blocks. We evaluate AV-CrossNet on multiple datasets, including LRS, VoxCeleb, and COG-MHEAR challenge. Evaluation results demonstrate that AV-CrossNet advances the state-of-the-art performance in all audiovisual tasks, even on untrained and mismatched datasets.

Anurag Kumar, Chinmay Bharti, Saikat Dutta et al.

Recent advancements in deep learning have demonstrated remarkable performance comparable to human capabilities across various supervised computer vision tasks. However, the prevalent assumption of having an extensive pool of training data encompassing all classes prior to model training often diverges from real-world scenarios, where limited data availability for novel classes is the norm. The challenge emerges in seamlessly integrating new classes with few samples into the training data, demanding the model to adeptly accommodate these additions without compromising its performance on base classes. To address this exigency, the research community has introduced several solutions under the realm of few-shot class incremental learning (FSCIL). In this study, we introduce an innovative FSCIL framework that utilizes language regularizer and subspace regularizer. During base training, the language regularizer helps incorporate semantic information extracted from a Vision-Language model. The subspace regularizer helps in facilitating the model's acquisition of nuanced connections between image and text semantics inherent to base classes during incremental training. Our proposed framework not only empowers the model to embrace novel classes with limited data, but also ensures the preservation of performance on base classes. To substantiate the efficacy of our approach, we conduct comprehensive experiments on three distinct FSCIL benchmarks, where our framework attains state-of-the-art performance.

Anastasia Kuznetsova, Anurag Kumar, Jennifer Drexler Fox et al.

Modern end-to-end speech recognition models show astonishing results in transcribing audio signals into written text. However, conventional data feeding pipelines may be sub-optimal for low-resource speech recognition, which still remains a challenging task. We propose an automated curriculum learning approach to optimize the sequence of training examples based on both the progress of the model while training and prior knowledge about the difficulty of the training examples. We introduce a new difficulty measure called compression ratio that can be used as a scoring function for raw audio in various noise conditions. The proposed method improves speech recognition Word Error Rate performance by up to 33% relative over the baseline system

Efthymios Tzinis, Yossi Adi, Vamsi Krishna Ithapu et al.

We present RemixIT, a simple yet effective self-supervised method for training speech enhancement without the need of a single isolated in-domain speech nor a noise waveform. Our approach overcomes limitations of previous methods which make them dependent on clean in-domain target signals and thus, sensitive to any domain mismatch between train and test samples. RemixIT is based on a continuous self-training scheme in which a pre-trained teacher model on out-of-domain data infers estimated pseudo-target signals for in-domain mixtures. Then, by permuting the estimated clean and noise signals and remixing them together, we generate a new set of bootstrapped mixtures and corresponding pseudo-targets which are used to train the student network. Vice-versa, the teacher periodically refines its estimates using the updated parameters of the latest student models. Experimental results on multiple speech enhancement datasets and tasks not only show the superiority of our method over prior approaches but also showcase that RemixIT can be combined with any separation model as well as be applied towards any semi-supervised and unsupervised domain adaptation task. Our analysis, paired with empirical evidence, sheds light on the inside functioning of our self-training scheme wherein the student model keeps obtaining better performance while observing severely degraded pseudo-targets.

Zhiqi Kang, Mostafa Sadeghi, Radu Horaud et al.

This paper investigates the impact of head movements on audio-visual speech enhancement (AVSE). Although being a common conversational feature, head movements have been ignored by past and recent studies: they challenge today's learning-based methods as they often degrade the performance of models that are trained on clean, frontal, and steady face images. To alleviate this problem, we propose to use robust face frontalization (RFF) in combination with an AVSE method based on a variational auto-encoder (VAE) model. We briefly describe the basic ingredients of the proposed pipeline and we perform experiments with a recently released audio-visual dataset. In the light of these experiments, and based on three standard metrics, namely STOI, PESQ and SI-SDR, we conclude that RFF improves the performance of AVSE by a considerable margin.

Jonah Casebeer, Jacob Donley, Daniel Wong et al.

Estimating a time-varying spatial covariance matrix for a beamforming algorithm is a challenging task, especially for wearable devices, as the algorithm must compensate for time-varying signal statistics due to rapid pose-changes. In this paper, we propose Neural Integrated Covariance Estimators for Beamformers, NICE-Beam. NICE-Beam is a general technique for learning how to estimate time-varying spatial covariance matrices, which we apply to joint speech enhancement and dereverberation. It is based on training a neural network module to non-linearly track and leverage scene information across time. We integrate our solution into a beamforming pipeline, which enables simple training, faster than real-time inference, and a variety of test-time adaptation options. We evaluate the proposed model against a suite of baselines in scenes with both stationary and moving microphones. Our results show that the proposed method can outperform a hand-tuned estimator, despite the hand-tuned estimator using oracle source separation knowledge.

Asutosh Pandey, Buye Xu, Anurag Kumar et al.

Deep neural networks are often coupled with traditional spatial filters, such as MVDR beamformers for effectively exploiting spatial information. Even though single-stage end-to-end supervised models can obtain impressive enhancement, combining them with a traditional beamformer and a DNN-based post-filter in a multistage processing provides additional improvements. In this work, we propose a two-stage strategy for multi-channel speech enhancement that does not require a traditional beamformer for additional performance. First, we propose a novel attentive dense convolutional network (ADCN) for estimating real and imaginary parts of complex spectrogram. ADCN obtains state-of-the-art results among single-stage models. Next, we use ADCN with a recently proposed triple-path attentive recurrent network (TPARN) for estimating waveform samples. The proposed strategy uses two insights; first, using different approaches in two stages; and second, using a stronger model in the first stage. We illustrate the efficacy of our strategy by evaluating multiple models in a two-stage approach with and without a traditional beamformer.

Ashutosh Pandey, Buye Xu, Anurag Kumar et al.

Deep neural networks (DNNs) are very effective for multichannel speech enhancement with fixed array geometries. However, it is not trivial to use DNNs for ad-hoc arrays with unknown order and placement of microphones. We propose a novel triple-path network for ad-hoc array processing in the time domain. The key idea in the network design is to divide the overall processing into spatial processing and temporal processing and use self-attention for spatial processing. Using self-attention for spatial processing makes the network invariant to the order and the number of microphones. The temporal processing is done independently for all channels using a recently proposed dual-path attentive recurrent network. The proposed network is a multiple-input multiple-output architecture that can simultaneously enhance signals at all microphones. Experimental results demonstrate the excellent performance of the proposed approach. Further, we present analysis to demonstrate the effectiveness of the proposed network in utilizing multichannel information even from microphones at far locations.

Ashutosh Pandey, Buye Xu, Anurag Kumar et al.

In this work, we propose a new model called triple-path attentive recurrent network (TPARN) for multichannel speech enhancement in the time domain. TPARN extends a single-channel dual-path network to a multichannel network by adding a third path along the spatial dimension. First, TPARN processes speech signals from all channels independently using a dual-path attentive recurrent network (ARN), which is a recurrent neural network (RNN) augmented with self-attention. Next, an ARN is introduced along the spatial dimension for spatial context aggregation. TPARN is designed as a multiple-input and multiple-output architecture to enhance all input channels simultaneously. Experimental results demonstrate the superiority of TPARN over existing state-of-the-art approaches.

Efthymios Tzinis, Yossi Adi, Vamsi K. Ithapu et al.

We propose RemixIT, a simple and novel self-supervised training method for speech enhancement. The proposed method is based on a continuously self-training scheme that overcomes limitations from previous studies including assumptions for the in-domain noise distribution and having access to clean target signals. Specifically, a separation teacher model is pre-trained on an out-of-domain dataset and is used to infer estimated target signals for a batch of in-domain mixtures. Next, we bootstrap the mixing process by generating artificial mixtures using permuted estimated clean and noise signals. Finally, the student model is trained using the permuted estimated sources as targets while we periodically update teacher's weights using the latest student model. Our experiments show that RemixIT outperforms several previous state-of-the-art self-supervised methods under multiple speech enhancement tasks. Additionally, RemixIT provides a seamless alternative for semi-supervised and unsupervised domain adaptation for speech enhancement tasks, while being general enough to be applied to any separation task and paired with any separation model.

Sangeeta Srivastava, Yun Wang, Andros Tjandra et al.

Representation learning from unlabeled data has been of major interest in artificial intelligence research. While self-supervised speech representation learning has been popular in the speech research community, very few works have comprehensively analyzed audio representation learning for non-speech audio tasks. In this paper, we propose a self-supervised audio representation learning method and apply it to a variety of downstream non-speech audio tasks. We combine the well-known wav2vec 2.0 framework, which has shown success in self-supervised learning for speech tasks, with parameter-efficient conformer architectures. Our self-supervised pre-training can reduce the need for labeled data by two-thirds. On the AudioSet benchmark, we achieve a mean average precision (mAP) score of 0.415, which is a new state-of-the-art on this dataset through audio-only self-supervised learning. Our fine-tuned conformers also surpass or match the performance of previous systems pre-trained in a supervised way on several downstream tasks. We further discuss the important design considerations for both pre-training and fine-tuning.

Kristen Grauman, Andrew Westbury, Eugene Byrne et al.

We introduce Ego4D, a massive-scale egocentric video dataset and benchmark suite. It offers 3,670 hours of daily-life activity video spanning hundreds of scenarios (household, outdoor, workplace, leisure, etc.) captured by 931 unique camera wearers from 74 worldwide locations and 9 different countries. The approach to collection is designed to uphold rigorous privacy and ethics standards with consenting participants and robust de-identification procedures where relevant. Ego4D dramatically expands the volume of diverse egocentric video footage publicly available to the research community. Portions of the video are accompanied by audio, 3D meshes of the environment, eye gaze, stereo, and/or synchronized videos from multiple egocentric cameras at the same event. Furthermore, we present a host of new benchmark challenges centered around understanding the first-person visual experience in the past (querying an episodic memory), present (analyzing hand-object manipulation, audio-visual conversation, and social interactions), and future (forecasting activities). By publicly sharing this massive annotated dataset and benchmark suite, we aim to push the frontier of first-person perception. Project page: https://ego4d-data.org/

Pranay Manocha, Buye Xu, Anurag Kumar

The perceptual task of speech quality assessment (SQA) is a challenging task for machines to do. Objective SQA methods that rely on the availability of the corresponding clean reference have been the primary go-to approaches for SQA. Clearly, these methods fail in real-world scenarios where the ground truth clean references are not available. In recent years, non-intrusive methods that train neural networks to predict ratings or scores have attracted much attention, but they suffer from several shortcomings such as lack of robustness, reliance on labeled data for training and so on. In this work, we propose a new direction for speech quality assessment. Inspired by human's innate ability to compare and assess the quality of speech signals even when they have non-matching contents, we propose a novel framework that predicts a subjective relative quality score for the given speech signal with respect to any provided reference without using any subjective data. We show that neural networks trained using our framework produce scores that correlate well with subjective mean opinion scores (MOS) and are also competitive to methods such as DNSMOS, which explicitly relies on MOS from humans for training networks. Moreover, our method also provides a natural way to embed quality-related information in neural networks, which we show is helpful for downstream tasks such as speech enhancement.

Yangyang Xia, Buye Xu, Anurag Kumar

Supervised speech enhancement relies on parallel databases of degraded speech signals and their clean reference signals during training. This setting prohibits the use of real-world degraded speech data that may better represent the scenarios where such systems are used. In this paper, we explore methods that enable supervised speech enhancement systems to train on real-world degraded speech data. Specifically, we propose a semi-supervised approach for speech enhancement in which we first train a modified vector-quantized variational autoencoder that solves a source separation task. We then use this trained autoencoder to further train an enhancement network using real-world noisy speech data by computing a triplet-based unsupervised loss function. Experiments show promising results for incorporating real-world data in training speech enhancement systems.

Ori Kabeli, Yossi Adi, Zhenyu Tang et al.

Deep neural networks have recently shown great success in the task of blind source separation, both under monaural and binaural settings. Although these methods were shown to produce high-quality separations, they were mainly applied under offline settings, in which the model has access to the full input signal while separating the signal. In this study, we convert a non-causal state-of-the-art separation model into a causal and real-time model and evaluate its performance under both online and offline settings. We compare the performance of the proposed model to several baseline methods under anechoic, noisy, and noisy-reverberant recording conditions while exploring both monaural and binaural inputs and outputs. Our findings shed light on the relative difference between causal and non-causal models when performing separation. Our stateful implementation for online separation leads to a minor drop in performance compared to the offline model; 0.8dB for monaural inputs and 0.3dB for binaural inputs while reaching a real-time factor of 0.65. Samples can be found under the following link: https://kwanum.github.io/sagrnnc-stream-results/.

Anurag Kumar, Yun Wang, Vamsi Krishna Ithapu et al.

Transfer learning is critical for efficient information transfer across multiple related learning problems. A simple, yet effective transfer learning approach utilizes deep neural networks trained on a large-scale task for feature extraction. Such representations are then used to learn related downstream tasks. In this paper, we investigate transfer learning capacity of audio representations obtained from neural networks trained on a large-scale sound event detection dataset. We build and evaluate these representations across a wide range of other audio tasks, via a simple linear classifier transfer mechanism. We show that such simple linear transfer is already powerful enough to achieve high performance on the downstream tasks. We also provide insights into the attributes of sound event representations that enable such efficient information transfer.

Pranay Manocha, Anurag Kumar, Buye Xu et al.

Subjective evaluations are critical for assessing the perceptual realism of sounds in audio-synthesis driven technologies like augmented and virtual reality. However, they are challenging to set up, fatiguing for users, and expensive. In this work, we tackle the problem of capturing the perceptual characteristics of localizing sounds. Specifically, we propose a framework for building a general purpose quality metric to assess spatial localization differences between two binaural recordings. We model localization similarity by utilizing activation-level distances from deep networks trained for direction of arrival (DOA) estimation. Our proposed metric (DPLM) outperforms baseline metrics on correlation with subjective ratings on a diverse set of datasets, even without the benefit of any human-labeled training data.

Avinash Mohan, Arpan Chattopadhyay, Shivam Vinayak Vatsa et al.

We consider a system of several collocated nodes sharing a time slotted wireless channel, and seek a MAC (medium access control) that (i) provides low mean delay, (ii) has distributed control (i.e., there is no central scheduler), and (iii) does not require explicit exchange of state information or control signals. The design of such MAC protocols must keep in mind the need for contention access at light traffic, and scheduled access in heavy traffic, leading to the long-standing interest in hybrid, adaptive MACs. Working in the discrete time setting, for the distributed MAC design, we consider a practical information structure where each node has local information and some common information obtained from overhearing. In this setting, "ZMAC" is an existing protocol that is hybrid and adaptive. We approach the problem via two steps (1) We show that it is sufficient for the policy to be "greedy" and "exhaustive". Limiting the policy to this class reduces the problem to obtaining a queue switching policy at queue emptiness instants. (2) Formulating the delay optimal scheduling as a POMDP (partially observed Markov decision process), we show that the optimal switching rule is Stochastic Largest Queue (SLQ). Using this theory as the basis, we then develop a practical distributed scheduler, QZMAC, which is also tunable. We implement QZMAC on standard off-the-shelf TelosB motes and also use simulations to compare QZMAC with the full-knowledge centralized scheduler, and with ZMAC. We use our implementation to study the impact of false detection while overhearing the common information, and the efficiency of QZMAC. Our simulation results show that the mean delay with QZMAC is close that of the full-knowledge centralized scheduler.

Panagiotis Tzirakis, Anurag Kumar, Jacob Donley

Multi-channel speech enhancement aims to extract clean speech from a noisy mixture using signals captured from multiple microphones. Recently proposed methods tackle this problem by incorporating deep neural network models with spatial filtering techniques such as the minimum variance distortionless response (MVDR) beamformer. In this paper, we introduce a different research direction by viewing each audio channel as a node lying in a non-Euclidean space and, specifically, a graph. This formulation allows us to apply graph neural networks (GNN) to find spatial correlations among the different channels (nodes). We utilize graph convolution networks (GCN) by incorporating them in the embedding space of a U-Net architecture. We use LibriSpeech dataset and simulate room acoustics data to extensively experiment with our approach using different array types, and number of microphones. Results indicate the superiority of our approach when compared to prior state-of-the-art method.

Anastasia Kuznetsova, Anurag Kumar, Francis M. Tyers

The Automated Speech Recognition (ASR) task has been a challenging domain especially for low data scenarios with few audio examples. This is the main problem in training ASR systems on the data from low-resource or marginalized languages. In this paper we present an approach to mitigate the lack of training data by employing Automated Curriculum Learning in combination with an adversarial bandit approach inspired by Reinforcement learning. The goal of the approach is to optimize the training sequence of mini-batches ranked by the level of difficulty and compare the ASR performance metrics against the random training sequence and discrete curriculum. We test our approach on a truly low-resource language and show that the bandit framework has a good improvement over the baseline transfer-learning model.

Ke Tan, Buye Xu, Anurag Kumar et al.

Most existing deep learning based binaural speaker separation systems focus on producing a monaural estimate for each of the target speakers, and thus do not preserve the interaural cues, which are crucial for human listeners to perform sound localization and lateralization. In this study, we address talker-independent binaural speaker separation with interaural cues preserved in the estimated binaural signals. Specifically, we extend a newly-developed gated recurrent neural network for monaural separation by additionally incorporating self-attention mechanisms and dense connectivity. We develop an end-to-end multiple-input multiple-output system, which directly maps from the binaural waveform of the mixture to those of the speech signals. The experimental results show that our proposed approach achieves significantly better separation performance than a recent binaural separation approach. In addition, our approach effectively preserves the interaural cues, which improves the accuracy of sound localization.

Anurag Kumar, Vamsi Krishna Ithapu

An important problem in machine auditory perception is to recognize and detect sound events. In this paper, we propose a sequential self-teaching approach to learning sounds. Our main proposition is that it is harder to learn sounds in adverse situations such as from weakly labeled and/or noisy labeled data, and in these situations a single stage of learning is not sufficient. Our proposal is a sequential stage-wise learning process that improves generalization capabilities of a given modeling system. We justify this method via technical results and on Audioset, the largest sound events dataset, our sequential learning approach can lead to up to 9% improvement in performance. A comprehensive evaluation also shows that the method leads to improved transferability of knowledge from previously trained models, thereby leading to improved generalization capabilities on transfer learning tasks.

Haytham M. Fayek, Anurag Kumar

Recognizing sounds is a key aspect of computational audio scene analysis and machine perception. In this paper, we advocate that sound recognition is inherently a multi-modal audiovisual task in that it is easier to differentiate sounds using both the audio and visual modalities as opposed to one or the other. We present an audiovisual fusion model that learns to recognize sounds from weakly labeled video recordings. The proposed fusion model utilizes an attention mechanism to dynamically combine the outputs of the individual audio and visual models. Experiments on the large scale sound events dataset, AudioSet, demonstrate the efficacy of the proposed model, which outperforms the single-modal models, and state-of-the-art fusion and multi-modal models. We achieve a mean Average Precision (mAP) of 46.16 on Audioset, outperforming prior state of the art by approximately +4.35 mAP (relative: 10.4%).

Anurag Kumar, Vamsi Krishna Ithapu

Weakly supervised learning algorithms are critical for scaling audio event detection to several hundreds of sound categories. Such learning models should not only disambiguate sound events efficiently with minimal class-specific annotation but also be robust to label noise, which is more apparent with weak labels instead of strong annotations. In this work, we propose a new framework for designing learning models with weak supervision by bridging ideas from sequential learning and knowledge distillation. We refer to the proposed methodology as SeCoST (pronounced Sequest) -- Sequential Co-supervision for training generations of Students. SeCoST incrementally builds a cascade of student-teacher pairs via a novel knowledge transfer method. Our evaluations on Audioset (the largest weakly labeled dataset available) show that SeCoST achieves a mean average precision of 0.383 while outperforming prior state of the art by a considerable margin.

Anurag Kumar, Ankit Shah, Bhiksha Raj et al.

In the last couple of years, weakly labeled learning has turned out to be an exciting approach for audio event detection. In this work, we introduce webly labeled learning for sound events which aims to remove human supervision altogether from the learning process. We first develop a method of obtaining labeled audio data from the web (albeit noisy), in which no manual labeling is involved. We then describe methods to efficiently learn from these webly labeled audio recordings. In our proposed system, WeblyNet, two deep neural networks co-teach each other to robustly learn from webly labeled data, leading to around 17% relative improvement over the baseline method. The method also involves transfer learning to obtain efficient representations