Shyamnath Gollakota

Bandhav Veluri, Collin Pernu, Ali Saffari et al.

We present NeuriCam, a novel deep learning-based system to achieve video capture from low-power dual-mode IoT camera systems. Our idea is to design a dual-mode camera system where the first mode is low-power (1.1 mW) but only outputs grey-scale, low resolution, and noisy video and the second mode consumes much higher power (100 mW) but outputs color and higher resolution images. To reduce total energy consumption, we heavily duty cycle the high power mode to output an image only once every second. The data for this camera system is then wirelessly sent to a nearby plugged-in gateway, where we run our real-time neural network decoder to reconstruct a higher-resolution color video. To achieve this, we introduce an attention feature filter mechanism that assigns different weights to different features, based on the correlation between the feature map and the contents of the input frame at each spatial location. We design a wireless hardware prototype using off-the-shelf cameras and address practical issues including packet loss and perspective mismatch. Our evaluations show that our dual-camera approach reduces energy consumption by 7x compared to existing systems. Further, our model achieves an average greyscale PSNR gain of 3.7 dB over prior single and dual-camera video super-resolution methods and 5.6 dB RGB gain over prior color propagation methods. Open-source code: https://github.com/vb000/NeuriCam.

Tuochao Chen, Qirui Wang, Bohan Wu et al.

Extracting the speech of participants in a conversation amidst interfering speakers and noise presents a challenging problem. In this paper, we introduce the novel task of target conversation extraction, where the goal is to extract the audio of a target conversation based on the speaker embedding of one of its participants. To accomplish this, we propose leveraging temporal patterns inherent in human conversations, particularly turn-taking dynamics, which uniquely characterize speakers engaged in conversation and distinguish them from interfering speakers and noise. Using neural networks, we show the feasibility of our approach on English and Mandarin conversation datasets. In the presence of interfering speakers, our results show an 8.19 dB improvement in signal-to-noise ratio for 2-speaker conversations and a 7.92 dB improvement for 2-4-speaker conversations. Code, dataset available at https://github.com/chentuochao/Target-Conversation-Extraction.

Bandhav Veluri, Justin Chan, Malek Itani et al.

We present the first neural network model to achieve real-time and streaming target sound extraction. To accomplish this, we propose Waveformer, an encoder-decoder architecture with a stack of dilated causal convolution layers as the encoder, and a transformer decoder layer as the decoder. This hybrid architecture uses dilated causal convolutions for processing large receptive fields in a computationally efficient manner while also leveraging the generalization performance of transformer-based architectures. Our evaluations show as much as 2.2-3.3 dB improvement in SI-SNRi compared to the prior models for this task while having a 1.2-4x smaller model size and a 1.5-2x lower runtime. We provide code, dataset, and audio samples: https://waveformer.cs.washington.edu/.

Bandhav Veluri, Benjamin N Peloquin, Bokai Yu et al.

Despite broad interest in modeling spoken dialogue agents, most approaches are inherently "half-duplex" -- restricted to turn-based interaction with responses requiring explicit prompting by the user or implicit tracking of interruption or silence events. Human dialogue, by contrast, is "full-duplex" allowing for rich synchronicity in the form of quick and dynamic turn-taking, overlapping speech, and backchanneling. Technically, the challenge of achieving full-duplex dialogue with LLMs lies in modeling synchrony as pre-trained LLMs do not have a sense of "time". To bridge this gap, we propose Synchronous LLMs for full-duplex spoken dialogue modeling. We design a novel mechanism to integrate time information into Llama3-8b so that they run synchronously with the real-world clock. We also introduce a training recipe that uses 212k hours of synthetic spoken dialogue data generated from text dialogue data to create a model that generates meaningful and natural spoken dialogue, with just 2k hours of real-world spoken dialogue data. Synchronous LLMs outperform state-of-the-art in dialogue meaningfulness while maintaining naturalness. Finally, we demonstrate the model's ability to participate in full-duplex dialogue by simulating interaction between two agents trained on different datasets, while considering Internet-scale latencies of up to 240 ms. Webpage: https://syncllm.cs.washington.edu/.

Bandhav Veluri, Malek Itani, Justin Chan et al.

Imagine being able to listen to the birds chirping in a park without hearing the chatter from other hikers, or being able to block out traffic noise on a busy street while still being able to hear emergency sirens and car honks. We introduce semantic hearing, a novel capability for hearable devices that enables them to, in real-time, focus on, or ignore, specific sounds from real-world environments, while also preserving the spatial cues. To achieve this, we make two technical contributions: 1) we present the first neural network that can achieve binaural target sound extraction in the presence of interfering sounds and background noise, and 2) we design a training methodology that allows our system to generalize to real-world use. Results show that our system can operate with 20 sound classes and that our transformer-based network has a runtime of 6.56 ms on a connected smartphone. In-the-wild evaluation with participants in previously unseen indoor and outdoor scenarios shows that our proof-of-concept system can extract the target sounds and generalize to preserve the spatial cues in its binaural output. Project page with code: https://semantichearing.cs.washington.edu

Maruchi Kim, Antonio Glenn, Bandhav Veluri et al.

Integrating cameras into wireless smart rings has been challenging due to size and power constraints. We introduce IRIS, the first wireless vision-enabled smart ring system for smart home interactions. Equipped with a camera, Bluetooth radio, inertial measurement unit (IMU), and an onboard battery, IRIS meets the small size, weight, and power (SWaP) requirements for ring devices. IRIS is context-aware, adapting its gesture set to the detected device, and can last for 16-24 hours on a single charge. IRIS leverages the scene semantics to achieve instance-level device recognition. In a study involving 23 participants, IRIS consistently outpaced voice commands, with a higher proportion of participants expressing a preference for IRIS over voice commands regarding toggling a device's state, granular control, and social acceptability. Our work pushes the boundary of what is possible with ring form-factor devices, addressing system challenges and opening up novel interaction capabilities.

Vidya Srinivas, Malek Itani, Tuochao Chen et al.

Models for low-latency, streaming applications could benefit from the knowledge capacity of larger models, but edge devices cannot run these models due to resource constraints. A possible solution is to transfer hints during inference from a large model running remotely to a small model running on-device. However, this incurs a communication delay that breaks real-time requirements and does not guarantee that both models will operate on the same data at the same time. We propose knowledge boosting, a novel technique that allows a large model to operate on time-delayed input during inference, while still boosting small model performance. Using a streaming neural network that processes 8 ms chunks, we evaluate different speech separation and enhancement tasks with communication delays of up to six chunks or 48 ms. Our results show larger gains where the performance gap between the small and large models is wide, demonstrating a promising method for large-small model collaboration for low-latency applications. Code, dataset, and audio samples available at https://knowledgeboosting.cs.washington.edu/.

Guilin Hu, Malek Itani, Tuochao Chen et al.

We introduce proactive hearing assistants that automatically identify and separate the wearer's conversation partners, without requiring explicit prompts. Our system operates on egocentric binaural audio and uses the wearer's self-speech as an anchor, leveraging turn-taking behavior and dialogue dynamics to infer conversational partners and suppress others. To enable real-time, on-device operation, we propose a dual-model architecture: a lightweight streaming model runs every 12.5 ms for low-latency extraction of the conversation partners, while a slower model runs less frequently to capture longer-range conversational dynamics. Results on real-world 2- and 3-speaker conversation test sets, collected with binaural egocentric hardware from 11 participants totaling 6.8 hours, show generalization in identifying and isolating conversational partners in multi-conversation settings. Our work marks a step toward hearing assistants that adapt proactively to conversational dynamics and engagement. More information can be found on our website: https://proactivehearing.cs.washington.edu/

Tuochao Chen, Bandhav Veluri, Hongyu Gong et al.

Dialogue models falter in noisy, multi-speaker environments, often producing irrelevant responses and awkward turn-taking. We present AV-Dialog, the first multimodal dialog framework that uses both audio and visual cues to track the target speaker, predict turn-taking, and generate coherent responses. By combining acoustic tokenization with multi-task, multi-stage training on monadic, synthetic, and real audio-visual dialogue datasets, AV-Dialog achieves robust streaming transcription, semantically grounded turn-boundary detection and accurate responses, resulting in a natural conversational flow. Experiments show that AV-Dialog outperforms audio-only models under interference, reducing transcription errors, improving turn-taking prediction, and enhancing human-rated dialogue quality. These results highlight the power of seeing as well as hearing for speaker-aware interaction, paving the way for {spoken} dialogue agents that perform {robustly} in real-world, noisy environments.

Bandhav Veluri, Malek Itani, Tuochao Chen et al.

In crowded settings, the human brain can focus on speech from a target speaker, given prior knowledge of how they sound. We introduce a novel intelligent hearable system that achieves this capability, enabling target speech hearing to ignore all interfering speech and noise, but the target speaker. A naive approach is to require a clean speech example to enroll the target speaker. This is however not well aligned with the hearable application domain since obtaining a clean example is challenging in real world scenarios, creating a unique user interface problem. We present the first enrollment interface where the wearer looks at the target speaker for a few seconds to capture a single, short, highly noisy, binaural example of the target speaker. This noisy example is used for enrollment and subsequent speech extraction in the presence of interfering speakers and noise. Our system achieves a signal quality improvement of 7.01 dB using less than 5 seconds of noisy enrollment audio and can process 8 ms of audio chunks in 6.24 ms on an embedded CPU. Our user studies demonstrate generalization to real-world static and mobile speakers in previously unseen indoor and outdoor multipath environments. Finally, our enrollment interface for noisy examples does not cause performance degradation compared to clean examples, while being convenient and user-friendly. Taking a step back, this paper takes an important step towards enhancing the human auditory perception with artificial intelligence. We provide code and data at: https://github.com/vb000/LookOnceToHear.

Victoria Ebert, Rishabh Singh, Tuochao Chen et al.

Conversational AI is constrained in many real-world settings where only one side of a dialogue can be recorded, such as telemedicine, call centers, and smart glasses. We formalize this as the one-sided conversation problem (1SC): inferring and learning from one side of a conversation. We study two tasks: (1) reconstructing the missing speaker's turns for real-time use cases, and (2) generating summaries from one-sided transcripts. Evaluating prompting and finetuned models on MultiWOZ, DailyDialog, and Candor with both human A/B testing and LLM-as-a-judge metrics, we find that access to one future turn and information about utterance length improves reconstruction, placeholder prompting helps to mitigate hallucination, and while large models generate promising reconstructions with prompting, smaller models require finetuning. Further, high-quality summaries can be generated without reconstructing missing turns. We present 1SC as a novel challenge and report promising results that mark a step toward privacy-aware conversational AI.

Wen Cheng, Tuochao Chen, Karim Helwani et al.

Edge devices such as smartwatches and smart glasses cannot continuously run even the smallest 100M-1B parameter language models due to power and compute constraints, yet cloud inference introduces multi-second latencies that break the illusion of a responsive assistant. We introduce micro language models ($μ$LMs): ultra-compact models (8M-30M parameters) that instantly generate the first 4-8 words of a contextually grounded response on-device, while a cloud model completes it; thus, masking the cloud latency. We show that useful language generation survives at this extreme scale with our models matching several 70M-256M-class existing models. We design a collaborative generation framework that reframes the cloud model as a continuator rather than a respondent, achieving seamless mid-sentence handoffs and structured graceful recovery via three error correction methods when the local opener goes wrong. Empirical results show that $μ$LMs can initiate responses that larger models complete seamlessly, demonstrating that orders-of-magnitude asymmetric collaboration is achievable and unlocking responsive AI for extremely resource-constrained devices. The model checkpoint and demo are available at https://github.com/Sensente/micro_language_model_swen_project.

Maruchi Kim, Rasya Fawwaz, Zhi Yang Lim et al.

Despite their ubiquity, wireless earbuds remain audio-centric due to size and power constraints. We present VueBuds, the first camera-integrated wireless earbuds for egocentric vision, capable of operating within stringent power and form-factor limits. Each VueBud embeds a camera into a Sony WF-1000XM3 to stream visual data over Bluetooth to a host device for on-device vision language model (VLM) processing. We show analytically and empirically that while each camera's field of view is partially occluded by the face, the combined binocular perspective provides comprehensive forward coverage. By integrating VueBuds with VLMs, we build an end-to-end system for real-time scene understanding, translation, visual reasoning, and text reading; all from low-resolution monochrome cameras drawing under 5mW through on-demand activation. Through online and in-person user studies with 90 participants, we compare VueBuds against smart glasses across 17 visual question-answering tasks, and show that our system achieves response quality on par with Ray-Ban Meta. Our work establishes low-power camera-equipped earbuds as a compelling platform for visual intelligence, bringing rapidly advancing VLM capabilities to one of the most ubiquitous wearable form factors.

Malek Itani, Tuochao Chen, Shyamnath Gollakota

Speech separation on hearable devices can enable transformative augmented and enhanced hearing capabilities. However, state-of-the-art speech separation networks cannot run in real-time on tiny, low-power neural accelerators designed for hearables, due to their limited compute capabilities. We present TF-MLPNet, the first speech separation network capable of running in real-time on such low-power accelerators while outperforming existing streaming models for blind speech separation and target speech extraction. Our network operates in the time-frequency domain, processing frequency sequences with stacks of fully connected layers that alternate along the channel and frequency dimensions, and independently processing the time sequence at each frequency bin using convolutional layers. Results show that our mixed-precision quantization-aware trained (QAT) model can process 6 ms audio chunks in real-time on the GAP9 processor, achieving a 3.5-4x runtime reduction compared to prior speech separation models.

Tuochao Chen, Nicholas Batchelder, Alisa Liu et al.

We introduce LlamaPIE, the first real-time proactive assistant designed to enhance human conversations through discreet, concise guidance delivered via hearable devices. Unlike traditional language models that require explicit user invocation, this assistant operates in the background, anticipating user needs without interrupting conversations. We address several challenges, including determining when to respond, crafting concise responses that enhance conversations, leveraging knowledge of the user for context-aware assistance, and real-time, on-device processing. To achieve this, we construct a semi-synthetic dialogue dataset and propose a two-model pipeline: a small model that decides when to respond and a larger model that generates the response. We evaluate our approach on real-world datasets, demonstrating its effectiveness in providing helpful, unobtrusive assistance. User studies with our assistant, implemented on Apple Silicon M2 hardware, show a strong preference for the proactive assistant over both a baseline with no assistance and a reactive model, highlighting the potential of LlamaPie to enhance live conversations.

Malek Itani, Ashton Graves, Sefik Emre Eskimez et al.

We present the first neural target speech extraction (TSE) system that uses human feedback for iterative refinement. Our approach allows users to mark specific segments of the TSE output, generating an edit mask. The refinement system then improves the marked sections while preserving unmarked regions. Since large-scale datasets of human-marked errors are difficult to collect, we generate synthetic datasets using various automated masking functions and train models on each. Evaluations show that models trained with noise power-based masking (in dBFS) and probabilistic thresholding perform best, aligning with human annotations. In a study with 22 participants, users showed a preference for refined outputs over baseline TSE. Our findings demonstrate that human-in-the-loop refinement is a promising approach for improving the performance of neural speech extraction.

Anran Wang, Maruchi Kim, Hao Zhang et al.

On-device directional hearing requires audio source separation from a given direction while achieving stringent human-imperceptible latency requirements. While neural nets can achieve significantly better performance than traditional beamformers, all existing models fall short of supporting low-latency causal inference on computationally-constrained wearables. We present DeepBeam, a hybrid model that combines traditional beamformers with a custom lightweight neural net. The former reduces the computational burden of the latter and also improves its generalizability, while the latter is designed to further reduce the memory and computational overhead to enable real-time and low-latency operations. Our evaluation shows comparable performance to state-of-the-art causal inference models on synthetic data while achieving a 5x reduction of model size, 4x reduction of computation per second, 5x reduction in processing time and generalizing better to real hardware data. Further, our real-time hybrid model runs in 8 ms on mobile CPUs designed for low-power wearable devices and achieves an end-to-end latency of 17.5 ms.

Justin Chan, Sharat Raju, Randall Bly et al.

Vascular anomalies, more colloquially known as birthmarks, affect up to 1 in 10 infants. Though many of these lesions self-resolve, some types can result in medical complications or disfigurement without proper diagnosis or management. Accurately diagnosing vascular anomalies is challenging for pediatricians and primary care physicians due to subtle visual differences and similarity to other pediatric dermatologic conditions. This can result in delayed or incorrect referrals for treatment. To address this problem, we developed a convolutional neural network (CNN) to automatically classify images of vascular anomalies and other pediatric skin conditions to aid physicians with diagnosis. We constructed a dataset of 21,681 clinical images, including data collected between 2002-2018 at Seattle Children's hospital as well as five dermatologist-curated online repositories, and built a taxonomy over vascular anomalies and other common pediatric skin lesions. The CNN achieved an average AUC of 0.9731 when ten-fold cross-validation was performed across a taxonomy of 12 classes. The classifier's average AUC and weighted F1 score was 0.9889 and 0.9732 respectively when evaluated on a previously unseen test set of six of these classes. Further, when used as an aid by pediatricians (n = 7), the classifier increased their average visual diagnostic accuracy from 73.10% to 91.67%. The classifier runs in real-time on a smartphone and has the potential to improve diagnosis of these conditions, particularly in resource-limited areas.

Anran Wang, Shyamnath Gollakota

Recent years have seen interest in device tracking and localization using acoustic signals. State-of-the-art acoustic motion tracking systems however do not achieve millimeter accuracy and require large separation between microphones and speakers, and as a result, do not meet the requirements for many VR/AR applications. Further, tracking multiple concurrent acoustic transmissions from VR devices today requires sacrificing accuracy or frame rate. We present MilliSonic, a novel system that pushes the limits of acoustic based motion tracking. Our core contribution is a novel localization algorithm that can provably achieve sub-millimeter 1D tracking accuracy in the presence of multipath, while using only a single beacon with a small 4-microphone array.Further, MilliSonic enables concurrent tracking of up to four smartphones without reducing frame rate or accuracy. Our evaluation shows that MilliSonic achieves 0.7mm median 1D accuracy and a 2.6mm median 3D accuracy for smartphones, which is 5x more accurate than state-of-the-art systems. MilliSonic enables two previously infeasible interaction applications: a) 3D tracking of VR headsets using the smartphone as a beacon and b) fine-grained 3D tracking for the Google Cardboard VR system using a small microphone array.

Saman Naderiparizi, Mehrdad Hessar, Vamsi Talla et al.

Wireless video streaming has traditionally been considered an extremely power-hungry operation. Existing approaches optimize the camera and communication modules individually to minimize their power consumption. However, the joint redesign and optimization of wireless communication as well as the camera is what that provides more power saving. We present an ultra-low-power wireless video streaming camera. To achieve this, we present a novel "analog" video backscatter technique that feeds analog pixels from the photo-diodes directly to the backscatter hardware, thereby eliminating power consuming hardware components such as ADCs and amplifiers. We prototype our wireless camera using off-the-shelf hardware and show that our design can stream video at up to 13 FPS and can operate up to a distance of 150 feet from the access point. Our COTS prototype consumes 2.36mW. Finally, to demonstrate the potential of our design, we built two proof-of-concept applications: video streaming for micro-robots and security cameras for face detection.

Rajalakshmi Nandakumar, Bryce Kellogg, Shyamnath Gollakota

This paper introduces the first wireless gesture recognition system that operates using existingWi-Fi signals and devices. To achieve this, we first identify limitations of existing wireless gesture recognition approaches that limit their applicability to Wi-Fi. We then introduce algorithms that can classify gestures using information that is readily available on Wi-Fi devices. We demonstrate the feasibility of our design using a prototype implementation on off-the-shelf Wi-Fi devices. Our results show that we can achieve a classification accuracy of 91% while classifying four gestures across six participants, without the need for per-participant training. Finally, we show the feasibility of gesture recognition in non-line-ofsight situations with the participants interacting with a Wi-Fi device placed in a backpack.