← Search

Trung Dang

6 accepted papers

2024

uaMix-MAE: Efficient Tuning of Pretrained Audio Transformers with Unsupervised Audio Mixtures

ICASSP 2024accepted

Masked Autoencoders (MAEs) learn rich low-level representations from unlabeled data but require substantial labeled data to effectively adapt to downstream tasks. Conversely, Instance Discrimination (ID) emphasizes high-level semantics, offering a potential solution to alleviate annotation requireme…

Cited by 0SourceScholar
2023

Optimality in Mean Estimation: Beyond Worst-Case, Beyond Sub-Gaussian, and Beyond $1+\alpha$ Moments

NeurIPS 2023poster

There is growing interest in improving our algorithmic understanding of fundamental statistical problems such as mean estimation, driven by the goal of understanding the fundamental limits of what we can extract from limited and valuable data. The state of the art results for mean estimation in $\ma…

Cited by 2SourcePDFScholar
2022

A Method to Reveal Speaker Identity in Distributed ASR Training, and How to Counter IT

ICASSP 2022accepted

End-to-end Automatic Speech Recognition (ASR) models are commonly trained over spoken utterances using optimization methods like Stochastic Gradient Descent (SGD). In distributed settings like Federated Learning, model training requires transmission of gradients over a network. In this work, we desi…

Cited by 0SourceScholar
2022

Training Robust Zero-Shot Voice Conversion Models with Self-Supervised Features

ICASSP 2022accepted

Unsupervised Zero-Shot Voice Conversion (VC) aims to modify the speaker characteristic of an utterance to match an unseen target speaker without relying on parallel training data. Recently, self-supervised learning of speech representation has been shown to produce useful linguistic units without us…

Cited by 0SourceScholar
2021

Revealing and Protecting Labels in Distributed Training

NeurIPS 2021poster

Distributed learning paradigms such as federated learning often involve transmission of model updates, or gradients, over a network, thereby avoiding transmission of private data. However, it is possible for sensitive information about the training data to be revealed from such gradients. Prior work…