Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 1907.04926

Synchronizing Audio-Visual Film Stimuli in Unity (version 5.5.1f1): Game Engines as a Tool for Research

Javier Sanz·Andreas Wulff-Abramsson·Carlos Aguilar-Paredes·Luis Emilio Bruni·Lydia Sanchez
View original source ↗Read source document ↗Export metadata ↓

Abstract

Unity is a software specifically designed for the development of video games. However, due to its programming possibilities and the polyvalence of its architecture, it can prove to be a versatile tool for stimuli presentation in research experiments. Nevertheless, it also has some limitations and conditions that need to be taken into account to ensure optimal performance in particular experimental situations. Such is the case if we want to use it in an experimental design that includes the acquisition of biometric signals synchronized with the broadcasting of video and audio in real time. In the present paper, we analyse how Unity (version 5.5.1f1) reacts in one such experimental design that requires the execution of audio-visual material. From the analysis of an experimental procedure in which the video was executed following the standard software specifications, we have detected the following problems desynchronization between the emission of the video and the audio; desynchronization between the temporary counter and the video; a delay in the execution of the screenshot; and depending on the encoding of the video a bad fluency in the video playback, which even though it maintains the total playback time, it causes Unity to freeze frames and proceed to compensate with little temporary jumps in the video. Finally, having detected all the problems, a compensation and verification process is designed to be able to work with audio-visual material in Unity (version 5.5.1f1) in an accurate way. We present a protocol for checks and compensations that allows solving these problems to ensure the execution of robust experiments in terms of reliability.

Explore related subjects

eess.AScs.MMcs.SDeess.IV

Keep this discovery

BibTeXRIS

Javier Sanz, Andreas Wulff-Abramsson, Carlos Aguilar-Paredes, Luis Emilio Bruni, Lydia Sanchez. 2019-07-05. Synchronizing Audio-Visual Film Stimuli in Unity (version 5.5.1f1): Game Engines as a Tool for Research. https://arxiv.org/abs/1907.04926

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Diarization Error Decomposition Under Pause Annotation Ambiguity

Speaker diarization evaluation is sensitive to ambiguity in pause annotation, which can inflate diarization error rate (DER) or obscure genuine model errors. We show that morphological closing, which has been used for pause-tolerant diarization evaluation, discards segment-level distinctions. Instead, we propose an exact, overlap-aware decomposition of standard DER into a pause-attributable component, consisting of errors compatible with pause filling, and a residual core component that can serve as a proxy for intrinsic diarization errors. The decomposition leaves DER unchanged, while the pause-attributable and core components vary monotonically with the pause threshold and eventually saturate. Experiments spanning synthetic transformations, annotation mismatch, cross-domain evaluation, and tight-boundary diarization show that the decomposition reveals error sources not apparent from standard DER.

eess.AS↗
arXiv2026

Less can be More: What Aspects of Speech Drive End-of-Turn Detection

In conversational AI, detecting when a speaker has finished talking is crucial for natural turn taking. While recent work incorporates semantics, the relative contribution of different modalities remains unclear. We present a controlled ablation of acoustic, prosodic, and semantic signals for streaming end of turn detection using a lightweight trimodal classifier. Under identical training conditions, the acoustic prosodic combination achieves the best balance of accuracy and latency, achieving utterance F1 of 0.93 with 7.8% false alarms at 400ms median latency. Adding text increases premature detections without improving performance. Feature space analysis confirms that prosodic features have the strongest class separability, while text representations overlap substantially. These findings suggest that turn-taking is primarily conveyed through intonation and silence patterns rather than semantic completeness, enabling faster and more reliable systems without expensive text inference.

eess.AS↗
arXiv2026

Downstream-Task-Aware Unified Source Separation

Task-aware unified source separation (TUSS) enables a single model to handle diverse separation tasks by conditioning on input prompts. However, conventional TUSS does not account for downstream task requirements, such as whether the enhanced speech will be used for human listening or automatic speech recognition (ASR). In this paper, we propose a prompt extension framework for TUSS that incorporates downstream task information into the input prompts and switches the loss function according to the given prompt during training, enabling outputs with different signal characteristics at inference time. Specifically, we introduce an ASR-dedicated prompt paired with a regularized loss function that reduces speech artifacts to improve ASR robustness, while the standard prompt is paired with the conventional SNR loss function. Experiments on the LibriSpeech and JNAS corpora demonstrate that the proposed joint-training scheme enables a single model to improve ASR performance over noisy input across a wide range of SNR conditions by selecting the ASR-dedicated prompt, while maintaining general speech enhancement quality when the standard prompt is used.

eess.AS↗