paper-with-me

Papers

MixRep: Hidden Representation Mixup for Low-Resource Speech Recognition

2023-10-27 · Jiamin Xie, John H. L. Hansen

In this paper, we present MixRep, a simple and effective data augmentation strategy based on mixup for low-resource ASR. MixRep interpolates the feature dimensions of hidden representations in the neural network that can be applied to both the acoustic feature input and the output of each layer, which generalizes the previous MixSpeech method. Further, we propose to combine the mixup with a regularization along the time axis of the input, which is shown as complementary. We apply MixRep to a Conformer encoder of an E2E LAS architecture trained with a joint CTC loss. We experiment on the WSJ dataset and subsets of the SWB dataset, covering reading and telephony conversational speech. Experimental results show that MixRep consistently outperforms other regularization methods for low-resource ASR. Compared to a strong SpecAugment baseline, MixRep achieves a +6.5\% and a +6.7\% relative WER reduction on the eval92 set and the Callhome part of the eval'2000 set.

📄 PDF Abstract BibTeX arXiv:2310.18450

Code (1)

jiamin1013/mixrep-espnet 공식 구현 pytorch

Tasks

Data Augmentationspeech-recognitionSpeech Recognition

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically
Mixup Mixup is a data augmentation technique that generates a weighted combination of random image pairs from the training data. Given two images and their ground truth labels:…

Similar Papers 제목 키워드 기반

AC-Mix: Self-Supervised Adaptation for Low-Resource Automatic Speech Recognition using Agnostic Contrastive Mixup

2024-10-18 · Carlos Carvalho, Alberto Abad

Self-supervised learning (SSL) leverages large amounts of unlabelled data to learn rich speech representations, fostering improvements in automatic speech recognition (ASR), even when only a small amount of labelled data…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Domain AdaptationGPU+3

Manifold Mixup: Better Representations by Interpolating Hidden States

2018-06-13 · ICLR 2019 5 · Vikas Verma, Alex Lamb, Christopher Beckham, Amir Najafi 외

Deep neural networks excel at learning the training data, but often provide incorrect and confident predictions when evaluated on slightly different test examples. This includes distribution shifts, outliers, and adversa…

Image Classification

FairMixRep : Self-supervised Robust Representation Learning for Heterogeneous Data with Fairness constraints

2020-10-07 · Souradip Chakraborty, Ekansh Verma, Saswata Sahoo, Jyotishka Datta

Representation Learning in a heterogeneous space with mixed variables of numerical and categorical types has interesting challenges due to its complex feature manifold. Moreover, feature learning in an unsupervised setup…

DecoderFairnessRepresentation Learning

STEMM: Self-learning with Speech-text Manifold Mixup for Speech Translation

2022-03-20 · ACL 2022 5 · Qingkai Fang, Rong Ye, Lei LI, Yang Feng 외

How to learn a better speech representation for end-to-end speech-to-text translation (ST) with limited labeled data? Existing techniques often attempt to transfer powerful machine translation (MT) capabilities to ST, bu…

Machine TranslationSpeech-to-TextSpeech-to-Text Translation

ShuffleMix: Improving Representations via Channel-Wise Shuffle of Interpolated Hidden States

2023-05-30 · KangJun Liu, Ke Chen, Lihua Guo, YaoWei Wang 외

Mixup style data augmentation algorithms have been widely adopted in various tasks as implicit network regularization on representation learning to improve model generalization, which can be achieved by a linear interpol…

BenchmarkingData AugmentationRepresentation Learning