Arobert: An asr robust pre-trained language model for spoken language understanding

C Wang, S Dai, Y Wang, F Yang, M Qiu… - … on Audio, Speech …, 2022 - ieeexplore.ieee.org
Spoken Language Understanding (SLU) aims to interpret the meanings of human speeches
in order to support various human-machine interaction systems. A key technique for SLU is …

Neural model reprogramming with similarity based mapping for low-resource spoken command recognition

H Yen, PJ Ku, CHH Yang, H Hu, SM Siniscalchi… - arXiv preprint arXiv …, 2021 - arxiv.org
In this study, we propose a novel adversarial reprogramming (AR) approach for low-
resource spoken command recognition (SCR), and build an AR-SCR system. The AR …

FastICARL: Fast incremental classifier and representation learning with efficient budget allocation in audio sensing applications

YD Kwon, J Chauhan, C Mascolo - arXiv preprint arXiv:2106.07268, 2021 - arxiv.org
Various incremental learning (IL) approaches have been proposed to help deep learning
models learn new tasks/classes continuously without forgetting what was learned previously …

FewJoint: few-shot learning for joint dialogue understanding

Y Hou, X Wang, C Chen, B Li, W Che… - International Journal of …, 2022 - Springer
Few-shot learning (FSL) is one of the key future steps in machine learning and raises a lot of
attention. In this paper, we focus on the FSL problem of dialogue understanding, which …

Multimodal approach for code-mixed speech sentiment classification

S Keshav, G Jyothish Lal, B Premjith - Advances in Signal Processing …, 2023 - Springer
Sentiment analysis is a natural language processing (NLP) technique used to classify a
statement into three polarities, namely, positive negative and neutral. Thus, speech …

On the efficiency of integrating self-supervised learning and meta-learning for user-defined few-shot keyword spotting

WT Kao, YK Wu, CP Chen, ZS Chen… - 2022 IEEE Spoken …, 2023 - ieeexplore.ieee.org
User-defined keyword spotting is a task to detect new spoken terms defined by users. This
can be viewed as a few-shot learning problem since it is unreasonable for users to define …

Meta-forecasting by combining global deep representations with local adaptation

R Grazzi, V Flunkert, D Salinas, T Januschowski… - arXiv preprint arXiv …, 2021 - arxiv.org
While classical time series forecasting considers individual time series in isolation, recent
advances based on deep learning showed that jointly learning from a large pool of related …

Meta auxiliary learning for low-resource spoken language understanding

Y Gao, J Feng, C Deng, S Zhang - arXiv preprint arXiv:2206.12774, 2022 - arxiv.org
Spoken language understanding (SLU) treats automatic speech recognition (ASR) and
natural language understanding (NLU) as a unified task and usually suffers from data …

[PDF][PDF] A Meta-Learning Approach for User-Defined Spoken Term Classification with Varying Classes and Examples.

Y Chen, T Ko, J Wang - Interspeech, 2021 - isca-archive.org
Recently we formulated a user-defined spoken term classification task as a few-shot
learning task and tackled the task using Model-Agnostic Meta-Learning (MAML) algorithm …