Audio-Visual Intelligence in Large Foundation Models

Audio-Visual Intelligence (AVI) has emerged as a central frontier in artificial intelligence, bridging auditory and visual modalities to enable machines that can perceive, generate, and interact in the multimodal real world. In the era of large foundation models, joint modeling of audio and vision has become increasingly crucial, i.e., not only for understanding but also for controllable generation and reasoning across dynamic, temporally grounded signals. Recent advances, such as Meta MovieGen and Google Veo-3, highlight the growing industrial and academic focus on unified audio-vision architectures that learn from massive multimodal data. However, despite rapid progress, the literature remains fragmented, spanning diverse tasks, inconsistent taxonomies, and heterogeneous evaluation practices that impede systematic comparison and knowledge integration. This survey provides the first comprehensive review of AVI through the lens of large foundation models. We establish a unified taxonomy covering the broad landscape of AVI tasks, ranging from understanding (e.g., speech recognition, sound localization) to generation (e.g., audio-driven video synthesis, video-to-audio) and interaction (e.g., dialogue, embodied, or agentic interfaces). We synthesize methodological foundations, including modality tokenization, cross-modal fusion, autoregressive and diffusion-based generation, large-scale pretraining, instruction alignment, and preference optimization. Furthermore, we curate representative datasets, benchmarks, and evaluation metrics, offering a structured comparison across task families and identifying open challenges in synchronization, spatial reasoning, controllability, and safety. By consolidating this rapidly expanding field into a coherent framework, this survey aims to serve as a foundational reference for future research on large-scale AVI.

MultiSense-Pneumo: A Multimodal Learning Framework for Pneumonia Screening in Resource-Constrained Settings

Pneumonia remains a leading global cause of morbidity and mortality, particularly in low resource settings where access to imaging, laboratory testing, and specialist care is limited. Clinical assessment relies on heterogeneous evidence, including symptoms, respiratory patterns, and chest imaging, making screening inherently multimodal. However, many existing computational approaches remain unimodal and focus primarily on radiographs. In this work, we present MultiSense-Pneumo, a multimodal framework for pneumonia oriented screening and triage support that integrates structured symptom descriptors, cough audio, spoken language, and chest radiographs. The system combines deterministic symptom triage, LightGBM based acoustic classification, domain adversarial radiograph analysis using ResNet 18, transformer based speech recognition, and an interpretable multimodal fusion operator. Each modality is transformed into a normalized risk signal and aggregated into a unified screening estimate, enabling transparent and modular decision support. MultiSense-Pneumo is designed for real world deployment under modest computational constraints and can operate fully offline on standard laptop class hardware, making it suitable for community health workers, rural clinics, and emergency response settings. Experimental results demonstrate robustness of the radiograph pathway under domain shifts, while highlighting limitations in minority class recall for acoustic signals. MultiSense-Pneumo is intended as a research prototype for screening and triage support rather than a clinically validated diagnostic system.

WhisperPipe: A Resource-Efficient Streaming Architecture for Real-Time Automatic Speech Recognition

Real-time automatic speech recognition (ASR) systems face a fundamental trade-off between transcription accuracy and computational efficiency, particularly when deploying large-scale transformer models like Whisper. Existing streaming approaches either sacrifice accuracy through aggressive chunking or incur prohibitive memory costs through unbounded context accumulation. We present WhisperPipe, a novel streaming architecture that achieves bounded memory consumption while maintaining transcription quality through three key innovations a hybrid Voice Activity Detection (VAD) pipeline combining Silero VAD with energy-based filtering to reduce false activations by 34%, a dynamic buffering mechanism with overlapping context windows that prevents information loss at segment boundaries, and an adaptive processing strategy that balances latency and accuracy based on speech characteristics. Evaluated on 2.5 hours of diverse audio data, WhisperPipe demonstrates a median end-to-end latency of 89ms (90th percentile: 142ms) while consuming 48% less peak GPU memory and 80.9% lower average GPU utilization compared to baseline Whisper implementations. The system maintains stable memory usage over extended sessions, with zero growth rate across 150-minute continuous operation. Comparative analysis against related work shows that WhisperPipe achieves competitive accuracy (WER within 2% of offline Whisper) while operating at 3-5x lower latency than existing streaming solutions. The architecture's modular design enables deployment across resource-constrained environments, from edge devices to cloud infrastructure. Our results demonstrate that careful architectural design can reconcile the competing demands of real-time responsiveness and model sophistication in production ASR systems.

Unrequited Emotions: Investigating the Gaps in Motivation and Practice in Speech Emotion Recognition Research

Critical analyses of emotion recognition technology have raised ethical concerns around task validity and potential downstream impacts, urging researchers to ensure alignment between their stated motivations and practice. However, these discussions have not adequately influenced or drawn from research on speech emotion recognition (SER). We address this gap by conducting a systematic survey of SER research to uncover what stated motivations drive this work and if they align with the datasets and emotions studied. We find that while SER research identifies appealing goals, such as well-situated voice-activated systems or healthcare applications, commonly-used datasets do not reflect these proposed deployment contexts, thus presenting a gap between motivations and research practices. We argue that such gaps engender ethical concerns, and that SER research should reassert itself with concrete use-cases to prevent misinterpretations, misuse, and downstream harms.

Evaluation of Automatic Speech Recognition Using Generative Large Language Models

Automatic Speech Recognition (ASR) is traditionally evaluated using Word Error Rate (WER), a metric that is insensitive to meaning. Embedding-based semantic metrics are better correlated with human perception, but decoder-based Large Language Models (LLMs) remain underexplored for this task. This paper evaluates their relevance through three approaches: (1) selecting the best hypothesis between two candidates, (2) computing semantic distance using generative embeddings, and (3) qualitative classification of errors. On the HATS dataset, the best LLMs achieve 92--94\% agreement with human annotators for hypothesis selection, compared to 63\% for WER, also outperforming semantic metrics. Embeddings from decoder-based LLMs show performance comparable to encoder models. Finally, LLMs offer a promising direction for interpretable and semantic ASR evaluation.

Identifying and typifying demographic unfairness in phoneme-level embeddings of self-supervised speech recognition models

Modern automatic speech recognition (ASR) systems have been observed to function better for certain speaker groups (SGs) than others, despite recent gains in overall performance. One potential impediment to progress towards fairer ASR is a more nuanced understanding of the types of modeling errors that speech encoder models make, and in particular the difference between the structure of embeddings for high-performance and low-performance SGs. This paper proposes a framework typifying two types of error that can occur in modeling phonemes in ASR systems: random error/high variance in phoneme embedding, vs systematic error/embedding bias. We find that training phoneme classification probes only on a single, typically disadvantaged SG, sometimes improves performance for that SG, which is evidence for the existence of SG-level bias in phoneme embeddings. On the other hand, we find that speakers and SGs with higher levels of phoneme variance are the same as those with worse phoneme prediction accuracy. We conclude that both types of error are present in phoneme embeddings and both are candidate causes for SG-level unfairness in ASR, though random error is likely a greater hindrance to fairness than systematic error. Furthermore, we find that finetuning encoder models using a fairness-enhancing algorithm (domain enhancing and adversarial training) changes neither the benefits of in-domain phoneme classification probe training, nor measured levels of random embedding error.

Advancing automatic speech recognition using feature fusion with self-supervised learning features: A case study on Fearless Steps Apollo corpus

Using self-supervised learning (SSL) models has significantly improved performance for downstream speech tasks, surpassing the capabilities of traditional hand-crafted features. This study investigates the amalgamation of SSL models, with the aim to leverage both their individual strengths and refine extracted features to achieve improved speech recognition models for naturalistic scenarios. Our research investigates the massive naturalistic Fearless Steps (FS) APOLLO resource, with particular focus on the FS Challenge (FSC) Phase-4 corpus, providing the inaugural analysis of this dataset. Additionally, we incorporate the CHiME-6 dataset to evaluate performance across diverse naturalistic speech scenarios. While exploring previously proposed Feature Refinement Loss and fusion methods, we found these methods to be less effective on the FSC Phase-4 corpus. To address this, we introduce a novel deep cross-attention (DCA) fusion method, designed to elevate performance, especially for the FSC Phase-4 corpus. Our objective is to foster creation of superior FS APOLLO community resources, catering to the diverse needs of researchers across various disciplines. The proposed solution achieves an absolute +1.1% improvement in WER, providing effective meta-data creation for the massive FS APOLLO community resource.

Do LLM Decoders Listen Fairly? Benchmarking How Language Model Priors Shape Bias in Speech Recognition

As pretrained large language models replace task-specific decoders in speech recognition, a critical question arises: do their text-derived priors make recognition fairer or more biased across demographic groups? We evaluate nine models spanning three architectural generations (CTC with no language model, encoder-decoder with an implicit LM, and LLM-based with an explicit pretrained decoder) on about 43,000 utterances across five demographic axes (ethnicity, accent, gender, age, first language) using Common Voice 24 and Meta's Fair-Speech, a controlled-prompt dataset that eliminates vocabulary confounds. On clean audio, three findings challenge assumptions: LLM decoders do not amplify racial bias (Granite-8B has the best ethnicity fairness, max/min WER = 2.28); Whisper exhibits pathological hallucination on Indian-accented speech with a non-monotonic insertion-rate spike to 9.62% at large-v3; and audio compression predicts accent fairness more than LLM scale. We then stress-test these findings under 12 acoustic degradation conditions (noise, reverberation, silence injection, chunk masking) across both datasets, totaling 216 inference runs. Severe degradation paradoxically compresses fairness gaps as all groups converge to high WER, but silence injection amplifies Whisper's accent bias up to 4.64x by triggering demographic-selective hallucination. Under masking, Whisper enters catastrophic repetition loops (86% of 51,797 insertions) while explicit-LLM decoders produce 38x fewer insertions with near-zero repetition; high-compression audio encoding (Q-former) reintroduces repetition pathology even in LLM decoders. These results suggest that audio encoder design, not LLM scaling, is the primary lever for equitable and robust speech recognition.

2nd of the 5th PVUW MeViS-Audio Track: ASR-SaSaSa2VA

Audio-based video object segmentation aims to locate and segment objects in videos conditioned on audio cues, requiring precise understanding of both appearance and motion. Recent audio-driven video segmentation methods extend MLLMs by fusing audio and visual features for end-to-end localization. Despite their promise, these approaches are computationally intensive, struggle with aligning temporal audio cues to dynamic video content, and depend on large paired audio-video datasets. To address these challenges, we present ASR-SaSaSa2VA, a resource-efficient framework for audio-guided video segmentation. The key idea is to convert audio inputs into textual motion descriptions via automatic speech recognition (ASR) models and then leverage pre-trained text-based referring video segmentation models (e.g., SaSaSa2VA) for pixel-level predictions. To further enhance robustness, we incorporate a no-target expression detection module, implemented by a fine-tuned audio-based MLLM, which filters out audio clips that do not refer to any target object. This design allows the system to exploit strong pre-trained models while effectively handling ambiguous or irrelevant audio inputs. Our approach achieves a final score of 80.7 in the 5th PVUW Challenge (MeViS-v2-Audio track), earning the second-place ranking.

UNet-Based Fusion and Exponential Moving Average Adaptation for Noise-Robust Speaker Recognition

The joint training of speech enhancement and speaker embedding networks for speaker recognition is widely adopted under noisy acoustic environments. While effective, this paradigm often fails to leverage the generalization and robustness benefits inherent in large-scale speech enhancement pre-training. Moreover, maintaining the speaker information in the denoised speech is not an explicit objective of the speech enhancement process. To address these limitations, we proposed a scalable \textbf{U}Net-based \textbf{F}usion framework (UF-EMA) that considers the noisy and enhanced speech as a multi-channel input, thereby enabling the speaker encoder to exploit speaker information effectively. In addition, an \textbf{E}xponential \textbf{M}oving \textbf{A}verage strategy is applied to a speaker encoder pre-trained on clean speech to mitigate overfitting and facilitate a smooth transition from clean to noisy conditions. Experimental results on multiple noise-contaminated test sets showcase the superiority of the proposed approach.