Temporarily-Aware Context Modelling using Generative Adversarial Networks for Speech Activity Detection
This work addresses speech activity detection, a key task in speech processing for applications like transcription and communication systems, with incremental improvements through a novel GAN-based framework.
The paper tackles speech activity detection by proposing a joint learning framework using generative adversarial networks with a temporal discriminator to ensure temporal consistency. The approach outperforms state-of-the-art methods on multiple public benchmarks including NIST OpenSAT'17, AMI Meeting, and HAVIC, and shows robustness across different languages, accents, and acoustic environments in cross-database evaluations.
This paper presents a novel framework for Speech Activity Detection (SAD). Inspired by the recent success of multi-task learning approaches in the speech processing domain, we propose a novel joint learning framework for SAD. We utilise generative adversarial networks to automatically learn a loss function for joint prediction of the frame-wise speech/ non-speech classifications together with the next audio segment. In order to exploit the temporal relationships within the input signal, we propose a temporal discriminator which aims to ensure that the predicted signal is temporally consistent. We evaluate the proposed framework on multiple public benchmarks, including NIST OpenSAT' 17, AMI Meeting and HAVIC, where we demonstrate its capability to outperform state-of-the-art SAD approaches. Furthermore, our cross-database evaluations demonstrate the robustness of the proposed approach across different languages, accents, and acoustic environments.