2023/09/29 by Samuel Pegg, Kai Li, Pegg, Samuel +3 · 5 citations
Computer Science · Engineering · Mathematics · Neuroscience · #Advanced Adaptive Filtering Techniques #Artificial intelligence #Audio and Speech Processing (eess.AS) #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Computer vision #Dimension (graph theory) #FOS: Computer and information sciences #FOS: Electrical engineering #Frequency domain #Hearing Loss and Rehabilitation #Inference #Machine learning #Mathematics #Separation (statistics) #Sound (cs.SD) #Speech and Audio Processing #Speech recognition #Time domain #Time–frequency analysis #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2309.17189
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2023/09/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Audio-visual speech separation methods aim to integrate different modalities to generate high-quality separated speech, thereby enhancing the performance of downstream tasks such as speech recognition. Most existing state-of-the-art (SOTA) models operate in the time domain. However, their overly simplistic approach to modeling acoustic features often necessitates larger and more computationally intensive models in order to achieve SOTA performance. In this paper, we present a novel time-frequency domain audio-visual speech separation method: Recurrent Time-Frequency Separation Network (RTFS-Net), which applies its algorithms on the complex time-frequency bins yielded by the Short-Time Fourier Transform. We model and capture the time and frequency dimensions of the audio independently using a multi-layered RNN along each dimension. Furthermore, we introduce a unique attention-based fusion technique for the efficient integration of audio and visual information, and a new mask separation approach that takes advantage of the intrinsic spectral nature of the acoustic features for a clearer separation. RTFS-Net outperforms the prior SOTA method in both inference speed and separation quality while reducing the number of parameters by 90% and MACs by 83%. This is the first time-frequency domain audio-visual speech separation method to outperform all contemporary time-domain counterparts.