Saved in:
Bibliographic Details
Main Authors: Huh, Mina, Ray, Ruchira, Karnei, Corey
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2303.00510
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910389274411008
author Huh, Mina
Ray, Ruchira
Karnei, Corey
author_facet Huh, Mina
Ray, Ruchira
Karnei, Corey
contents Data augmentations are known to improve robustness in speech-processing tasks. In this study, we summarize and compare different data augmentation strategies using S3PRL toolkit. We explore how HuBERT and wav2vec perform using different augmentation techniques (SpecAugment, Gaussian Noise, Speed Perturbation) for Phoneme Recognition (PR) and Automatic Speech Recognition (ASR) tasks. We evaluate model performance in terms of phoneme error rate (PER) and word error rate (WER). From the experiments, we observed that SpecAugment slightly improves the performance of HuBERT and wav2vec on the original dataset. Also, we show that models trained using the Gaussian Noise and Speed Perturbation dataset are more robust when tested with augmented test sets.
format Preprint
id arxiv_https___arxiv_org_abs_2303_00510
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A Comparison of Speech Data Augmentation Methods Using S3PRL Toolkit
Huh, Mina
Ray, Ruchira
Karnei, Corey
Sound
Artificial Intelligence
Audio and Speech Processing
Data augmentations are known to improve robustness in speech-processing tasks. In this study, we summarize and compare different data augmentation strategies using S3PRL toolkit. We explore how HuBERT and wav2vec perform using different augmentation techniques (SpecAugment, Gaussian Noise, Speed Perturbation) for Phoneme Recognition (PR) and Automatic Speech Recognition (ASR) tasks. We evaluate model performance in terms of phoneme error rate (PER) and word error rate (WER). From the experiments, we observed that SpecAugment slightly improves the performance of HuBERT and wav2vec on the original dataset. Also, we show that models trained using the Gaussian Noise and Speed Perturbation dataset are more robust when tested with augmented test sets.
title A Comparison of Speech Data Augmentation Methods Using S3PRL Toolkit
topic Sound
Artificial Intelligence
Audio and Speech Processing
url https://arxiv.org/abs/2303.00510