no code implementations • 25 Apr 2023 • Tanmay Khandelwal, Rohan Kumar Das, Andrew Koh, Eng Siong Chng
Stage-1 of our proposed framework focuses on audio-tagging (AT), which assists the sound event detection (SED) system in Stage-2.
no code implementations • 12 Sep 2022 • Andrew Koh, Sivakorn Sanguanmoo
We use these results to solve the problem of a designer with (i) nonlinear preferences over DM's stopping times: optimal structures have a block structure such that DM's indifference times coincide with the support of her stopping time; and (ii) preferences over actions and stopping times when they are additive or supermodular.
no code implementations • 29 Jun 2022 • Andrew Koh, Eng Siong Chng
In this paper, we tackle the new Language-Based Audio Retrieval task proposed in DCASE 2022.
no code implementations • 4 Jun 2022 • Andrew Koh, Soham Tiwari, Chng Eng Siong
In this paper, we propose an algorithm, Epochal Difficult Captions, to supplement the training of any model for the Automated Audio Captioning task.
no code implementations • 10 Aug 2021 • Andrew Koh, Fuzhao Xue, Eng Siong Chng
In this paper, we examine the use of Transfer Learning using Pretrained Audio Neural Networks (PANNs), and propose an architecture that is able to better leverage the acoustic features provided by PANNs for the Automated Audio Captioning Task.
1 code implementation • Findings of the Association for Computational Linguistics 2020 • Boon Peng Yap, Andrew Koh, Eng Siong Chng
Domain adaptation or transfer learning using pre-trained language models such as BERT has proven to be an effective approach for many natural language processing tasks.