UNSUPERVISED SPEECH RECOGNITION WITH N-SKIPGRAM AND POSITIONAL UNIGRAM MATCHING

Liming Wang; Mark Hasegawa-Johnson; Chang D. Yoo

doi:10.1109/ICASSP48485.2024.10446327

UNSUPERVISED SPEECH RECOGNITION WITH N-SKIPGRAM AND POSITIONAL UNIGRAM MATCHING

Liming Wang, Mark Hasegawa-Johnson, Chang D. Yoo

Research output: Contribution to journal › Conference article › peer-review

Abstract

Training unsupervised speech recognition systems presents challenges due to GAN-associated instability, misalignment between speech and text, and significant memory demands. To tackle these challenges, we introduce a novel ASR system, ESPUM. This system harnesses the power of lower-order N-skipgrams (up to N = 3) combined with positional unigram statistics gathered from a small batch of samples. Evaluated on the TIMIT benchmark, our model showcases competitive performance in ASR and phoneme segmentation tasks. Access our publicly available code at https://github.com/lwang114/GraphUnsupASR.

Original language	English (US)
Pages (from-to)	10936-10940
Number of pages	5
Journal	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
DOIs	https://doi.org/10.1109/ICASSP48485.2024.10446327
State	Published - 2024
Event	2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - Seoul, Korea, Republic of Duration: Apr 14 2024 → Apr 19 2024

Keywords

acoustic unit discovery
self-supervised speech processing
speech recognition
unsupervised phoneme segmentation

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Online availability

10.1109/ICASSP48485.2024.10446327

Library availability

Discover UIUC Full Text

Cite this

@article{a1ed5e52f75b43ec8bbb496d6f9700af,

title = "UNSUPERVISED SPEECH RECOGNITION WITH N-SKIPGRAM AND POSITIONAL UNIGRAM MATCHING",

abstract = "Training unsupervised speech recognition systems presents challenges due to GAN-associated instability, misalignment between speech and text, and significant memory demands. To tackle these challenges, we introduce a novel ASR system, ESPUM. This system harnesses the power of lower-order N-skipgrams (up to N = 3) combined with positional unigram statistics gathered from a small batch of samples. Evaluated on the TIMIT benchmark, our model showcases competitive performance in ASR and phoneme segmentation tasks. Access our publicly available code at https://github.com/lwang114/GraphUnsupASR.",

keywords = "acoustic unit discovery, self-supervised speech processing, speech recognition, unsupervised phoneme segmentation",

author = "Liming Wang and Mark Hasegawa-Johnson and Yoo, {Chang D.}",

note = "Publisher Copyright: {\textcopyright}2024 IEEE.; 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 ; Conference date: 14-04-2024 Through 19-04-2024",

year = "2024",

doi = "10.1109/ICASSP48485.2024.10446327",

language = "English (US)",

pages = "10936--10940",

journal = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

issn = "1520-6149",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - UNSUPERVISED SPEECH RECOGNITION WITH N-SKIPGRAM AND POSITIONAL UNIGRAM MATCHING

AU - Wang, Liming

AU - Hasegawa-Johnson, Mark

AU - Yoo, Chang D.

PY - 2024

Y1 - 2024

N2 - Training unsupervised speech recognition systems presents challenges due to GAN-associated instability, misalignment between speech and text, and significant memory demands. To tackle these challenges, we introduce a novel ASR system, ESPUM. This system harnesses the power of lower-order N-skipgrams (up to N = 3) combined with positional unigram statistics gathered from a small batch of samples. Evaluated on the TIMIT benchmark, our model showcases competitive performance in ASR and phoneme segmentation tasks. Access our publicly available code at https://github.com/lwang114/GraphUnsupASR.

AB - Training unsupervised speech recognition systems presents challenges due to GAN-associated instability, misalignment between speech and text, and significant memory demands. To tackle these challenges, we introduce a novel ASR system, ESPUM. This system harnesses the power of lower-order N-skipgrams (up to N = 3) combined with positional unigram statistics gathered from a small batch of samples. Evaluated on the TIMIT benchmark, our model showcases competitive performance in ASR and phoneme segmentation tasks. Access our publicly available code at https://github.com/lwang114/GraphUnsupASR.

KW - acoustic unit discovery

KW - self-supervised speech processing

KW - speech recognition

KW - unsupervised phoneme segmentation

UR - http://www.scopus.com/inward/record.url?scp=105001548137&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=105001548137&partnerID=8YFLogxK

U2 - 10.1109/ICASSP48485.2024.10446327

DO - 10.1109/ICASSP48485.2024.10446327

M3 - Conference article

AN - SCOPUS:105001548137

SN - 1520-6149

SP - 10936

EP - 10940

JO - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

JF - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

T2 - 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024

Y2 - 14 April 2024 through 19 April 2024

ER -

UNSUPERVISED SPEECH RECOGNITION WITH N-SKIPGRAM AND POSITIONAL UNIGRAM MATCHING

Abstract

Keywords

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this