# An Efficient and Streaming Audio Visual Active Speaker Detection System

arXiv: https://arxiv.org/abs/2409.09018 · HTML: https://mlsanity.com/paper/2409.09018/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 14 references that are in the map.
- First posted on arXiv 13 Sep 2024.

## Builds on (most cited first, 14 of 14)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [A Closer Look at Spatiotemporal Convolutions for Action Recognition](https://mlsanity.com/paper/1711.11248/) (2017)
- [Efficiently Scaling Transformer Inference](https://mlsanity.com/paper/2211.05102/) (2022)
- [Is Someone Speaking? Exploring Long-term Temporal Features for Audio-visual Active Speaker Detection](https://mlsanity.com/paper/2107.06592/) (2021)
- [AVA-ActiveSpeaker: An Audio-Visual Dataset for Active Speaker Detection](https://mlsanity.com/paper/1901.01342/) (2019)
- [Active Speakers in Context](https://mlsanity.com/paper/2005.09812/) (2020)
- [A Light Weight Model for Active Speaker Detection](https://mlsanity.com/paper/2303.04439/) (2023)
- [End-to-End Active Speaker Detection](https://mlsanity.com/paper/2203.14250/) (2022)
- [LoCoNet: Long-Short Context Network for Active Speaker Detection](https://mlsanity.com/paper/2301.08237/) (2023)
- [Cross-modal Supervision for Learning Active Speaker Detection in Video](https://mlsanity.com/paper/1603.08907/) (2016)
- [TalkNCE: Improving Active Speaker Detection with Talk-Aware Contrastive Learning](https://mlsanity.com/paper/2309.12306/) (2023)
- [Best of Both Worlds: Multi-task Audio-Visual Automatic Speech Recognition and Active Speaker Detection](https://mlsanity.com/paper/2205.05206/) (2022)

## Built on by (most cited first, 1 of 1)

- [CAST-ASD: Causal Attention over Speakers and Time for Online Active Speaker Detection](https://mlsanity.com/paper/2610.14676/) (2026)

## Abstract (arXiv)

This paper delves into the challenging task of Active Speaker Detection (ASD), where the system needs to determine in real-time whether a person is speaking or not in a series of video frames. While previous works have made significant strides in improving network architectures and learning effective representations for ASD, a critical gap exists in the exploration of real-time system deployment. Existing models often suffer from high latency and memory usage, rendering them impractical for immediate applications. To bridge this gap, we present two scenarios that address the key challenges posed by real-time constraints. First, we introduce a method to limit the number of future context frames utilized by the ASD model. By doing so, we alleviate the need for processing the entire sequence of future frames before a decision is made, significantly reducing latency. Second, we propose a more stringent constraint that limits the total number of past frames the model can access during inference. This tackles the persistent memory issues associated with running streaming ASD systems. Beyond these theoretical frameworks, we conduct extensive experiments to validate our approach. Our results demonstrate that constrained transformer models can achieve performance comparable to or even better than state-of-the-art recurrent models, such as uni-directional GRUs, with a significantly reduced number of context frames. Moreover, we shed light on the temporal memory requirements of ASD systems, revealing that larger past context has a more profound impact on accuracy than future context. When profiling on a CPU we find that our efficient architecture is memory bound by the amount of past context it can use and that the compute cost is negligible as compared to the memory cost.