arXiv subjects

Bill Wang

Publications and source records attributed to Bill Wang.

6 recordsLinked to original sources

arXiv2026

VoiceCodeBench: Evaluating Exact Structured-Token Recovery in Automatic Speech Recognition

Automatic speech recognition (ASR) systems are commonly evaluated with word error rate (WER), yet many voice workflows depend on exact written values for identifiers, paths, and measured quantities. A transcript can appear fluent and achieve low WER while corrupting a value that a downstream system must parse, store, or execute. We introduce VoiceCodeBench, a benchmark for evaluating exact structured-token recovery in English ASR. It contains 300 human-recorded workplace segments spanning eight workflow domains and 1,482 audited target entities across 26 entity types, each with a canonical written form recoverable from the audio. Under a raw-audio-only protocol, systems receive audio bytes without additional context or metadata. Alongside WER, we evaluate Canonical Token/Entity Match (CTEM), Task Success Rate (TSR), and per-type exact recovery. Across 12 baseline ASR systems, lower WER generally corresponded to better structured-token recovery but did not fully determine it: Spearman correlations were -0.73 for both WER versus CTEM and WER versus TSR. The strongest baseline by TSR reached only 68.7%, leaving nearly one third of recordings with at least one unrecovered workflow-critical value. These results show that entity-sensitive metrics are needed to assess whether ASR output preserves exact values that production systems must parse, route, store, compare, or execute.

cs.CL↗

arXiv2026

VocalAffectBench: Evaluating Vocal Emotion Recognition in AI Audio Models

Voice products increasingly need affective cues that are present in speech but absent from transcripts. We introduce VocalAffectBench, a public, test-only benchmark for evaluating whether AI audio models can identify expressed vocal emotion from raw audio. The benchmark contains 273 human-recorded English WAV clips from 51 speaker accounts totaling 1.95 hours across seven labels: angry, disgusted, fearful, happy, neutral, sad, and surprised, with 39 clips per class. All baselines are evaluated from audio alone, without transcripts or contextual metadata. Across six released baselines, average accuracy is 35.5%. The strongest baseline, gemini_3_5_flash, reaches 46.5% on the seven-way task, above the 14.3% random baseline but far from robust emotion recognition. A secondary valence-bucket analysis maps labels into positive, neutral, and negative classes, excluding surprised because its valence is ambiguous. Aggregate accuracy under this coarser view is 50.9%. Performance is highly uneven across classes. By recall, neutral is identified most reliably at 75.6% averaged across baselines, while surprised and fearful reach only 10.7% and 15.4%, respectively. These results show that the evaluated baselines can extract some affective signal from speech, but discrete expressed-emotion recognition remains fragile, especially for non-neutral emotions that are often most important in voice agent workflows.

cs.CL↗

arXiv2026

Random Matching Markets with Correlated Preferences

In the Gale-Shapley model of two-sided matching, it is well known that for generic preferences, the outcomes for each side can vary dramatically in the male-optimal vs. female-optimal stable matchings. In this paper, we show that under a widely used characterization of similarity in rankings, even a weak correlation in preferences guarantees assortative matching with high probability as the market size tends to infinity. It follows that the men's average ranking of women and the women's average ranking of men are asymptotically equivalent in all stable matchings with high probability, as long as the market imbalance is not too extreme.

econ.TH↗

arXiv2026

A Design Space for Live Music Agents

Live music provides a uniquely rich setting for studying creativity and interaction due to its spontaneous nature. The pursuit of live music agents--intelligent systems supporting real-time music performance and interaction--has captivated researchers across HCI, AI, and computer music for decades, and recent advancements in AI suggest unprecedented opportunities to evolve their design. However, the interdisciplinary nature of music has led to fragmented development across research communities, hindering effective communication and collaborative progress. In this work, we bring together perspectives from these diverse fields to map the current landscape of live music agents. Based on our analysis of 184 systems across both academic literature and video, we develop a comprehensive design space that categorizes dimensions spanning usage contexts, interactions, technologies, and ecosystems. By highlighting trends and gaps in live music agents, our design space offers researchers, designers, and musicians a structured lens to understand existing systems and shape future directions in real-time human-AI music co-creation. We release our annotated systems as a living artifact at https://live-music-agents.github.io.

cs.HC↗

arXiv2022

A Simulation Study of Passing Drivers' Responses to the Autonomous Truck-Mounted Attenuator System in Road Maintenance

The Autonomous Truck-Mounted Attenuator (ATMA) system is a lead-follower vehicle system based on autonomous driving and connected vehicle technologies. The lead truck performs maintenance tasks on the road, and the unmanned follower truck alerts passing vehicles about the moving work zone and protects workers and the equipment. While the ATMA has been under testing by transportation maintenance and operations agencies recently, a simulator-based testing capability is a supplement, especially if human subjects are involved. This paper aims to discover how passing drivers perceive, understand, and react to the ATMA system in road maintenance. With the driving simulator developed for this ATMA study, the paper performed a simulation study wherein a screen-based eye tracker collected sixteen subjects' gaze points and pupil diameters. Data analysis evidenced the change in subjects' visual attention patterns while passing the ATMA. On average, the ATMA starts to attract subjects' attention from 500 ft behind the follower truck. Most (87.50%) understood the follower truck's protection purpose, and many (60%) reasoned the association between the two trucks. Nevertheless, nearly half of the participants (43.75%) did not recognize that ATMA is a connected autonomous vehicle system. While all subjects safely changed lanes and attempted to pass the slow-moving ATMA, their inadequate understanding of the ATMA is a potential risk, like cutting into the ATAM. Results implied that transportation maintenance and operations agencies should consider this in establishing the deployment guidance.

cs.RO↗

arXiv2022

A Multi-tasking Model of Speaker-Keyword Classification for Keeping Human in the Loop of Drone-assisted Inspection

Audio commands are a preferred communication medium to keep inspectors in the loop of civil infrastructure inspection performed by a semi-autonomous drone. To understand job-specific commands from a group of heterogeneous and dynamic inspectors, a model must be developed cost-effectively for the group and easily adapted when the group changes. This paper is motivated to build a multi-tasking deep learning model that possesses a Share-Split-Collaborate architecture. This architecture allows the two classification tasks to share the feature extractor and then split subject-specific and keyword-specific features intertwined in the extracted features through feature projection and collaborative training. A base model for a group of five authorized subjects is trained and tested on the inspection keyword dataset collected by this study. The model achieved a 95.3% or higher mean accuracy in classifying the keywords of any authorized inspectors. Its mean accuracy in speaker classification is 99.2%. Due to the richer keyword representations that the model learns from the pooled training data, adapting the base model to a new inspector requires only a little training data from that inspector, like five utterances per keyword. Using the speaker classification scores for inspector verification can achieve a success rate of at least 93.9% in verifying authorized inspectors and 76.1% in detecting unauthorized ones. Further, the paper demonstrates the applicability of the proposed model to larger-size groups on a public dataset. This paper provides a solution to addressing challenges facing AI-assisted human-robot interaction, including worker heterogeneity, worker dynamics, and job heterogeneity.

cs.SD↗