Exploiting A Quantum Multiple Kernel Learning Approach For Low-Resource Spoken Command Recognition

Xianyan Fu; Xiao Lei Zhang; Chao Han Huck Yang; Jun Qi

doi:10.1109/ICASSP48485.2024.10448120

Exploiting A Quantum Multiple Kernel Learning Approach For Low-Resource Spoken Command Recognition

Xianyan Fu, Xiao Lei Zhang^*, Chao Han Huck Yang, Jun Qi

^*Corresponding author for this work

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

Abstract

We propose a theoretical analysis of quantum projection learning (QPL) that employs multiple kernels, highlighting its advantages through representation error analysis. Building upon previous studies that utilized a single quantum kernel-based method, we further investigate a quantum projection framework that incorporates multiple Gaussian kernels for low-resource spoken command recognition. Our empirical results align with our theoretical insights, suggesting that methods based on multiple kernels can further enhance the performance of QPL. By leveraging the quantum-to-classical projected output embeddings, we integrate this with a prototypical network for acoustic modeling. When evaluated using Arabic, Chuvash, Irish, and Lithuanian low-resource speech from CommonVoice, our proposed method surpasses the recurrent neural network and single kernel-based classifier baselines by an average of +5.28%.

Original language	English
Title of host publication	2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - Proceedings
Publisher	IEEE
Pages	12931-12935
Number of pages	5
ISBN (Electronic)	9798350344851
ISBN (Print)	9798350344868
DOIs	https://doi.org/10.1109/ICASSP48485.2024.10448120
Publication status	Published - Apr 2024
Event	2024 49th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - COEX, Seoul, Korea, Republic of Duration: 14 Apr 2024 → 19 Apr 2024 https://2024.ieeeicassp.org/ https://2024.ieeeicassp.org/program-schedule/ https://ieeexplore.ieee.org/xpl/conhome/10445798/proceeding

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)	1520-6149
ISSN (Electronic)	2379-190X

Conference

Conference	2024 49th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024
Abbreviated title	ICASSP 2024
Country/Territory	Korea, Republic of
City	Seoul
Period	14/04/24 → 19/04/24
Internet address	https://2024.ieeeicassp.org/ https://2024.ieeeicassp.org/program-schedule/ https://ieeexplore.ieee.org/xpl/conhome/10445798/proceeding

Scopus Subject Areas

Software
Signal Processing
Electrical and Electronic Engineering

User-Defined Keywords

low-resource speech classification
multiple kernel learning
quantum kernel projection

Access to Document

10.1109/ICASSP48485.2024.10448120

Cite this

Fu, X., Zhang, X. L., Yang, C. H. H., & Qi, J. (2024). Exploiting A Quantum Multiple Kernel Learning Approach For Low-Resource Spoken Command Recognition. In 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - Proceedings (pp. 12931-12935). (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). IEEE. https://doi.org/10.1109/ICASSP48485.2024.10448120

Fu, Xianyan ; Zhang, Xiao Lei ; Yang, Chao Han Huck et al. / Exploiting A Quantum Multiple Kernel Learning Approach For Low-Resource Spoken Command Recognition. 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - Proceedings. IEEE, 2024. pp. 12931-12935 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{a903e6ee4bd04728a921fb1e8dd8a6a6,

title = "Exploiting A Quantum Multiple Kernel Learning Approach For Low-Resource Spoken Command Recognition",

abstract = "We propose a theoretical analysis of quantum projection learning (QPL) that employs multiple kernels, highlighting its advantages through representation error analysis. Building upon previous studies that utilized a single quantum kernel-based method, we further investigate a quantum projection framework that incorporates multiple Gaussian kernels for low-resource spoken command recognition. Our empirical results align with our theoretical insights, suggesting that methods based on multiple kernels can further enhance the performance of QPL. By leveraging the quantum-to-classical projected output embeddings, we integrate this with a prototypical network for acoustic modeling. When evaluated using Arabic, Chuvash, Irish, and Lithuanian low-resource speech from CommonVoice, our proposed method surpasses the recurrent neural network and single kernel-based classifier baselines by an average of +5.28%.",

keywords = "low-resource speech classification, multiple kernel learning, quantum kernel projection",

author = "Xianyan Fu and Zhang, {Xiao Lei} and Yang, {Chao Han Huck} and Jun Qi",

note = "This work was supported in part by the National Science Foundation of China (NSFC) under Grant 62176211, and in part by the Project of the Science, Technology, and Innovation Commission of Shenzhen Municipality, China under Grant JCYJ20210324143006016 and JSGG20210802152546026. Publisher Copyright: {\textcopyright} 2024 IEEE.; 2024 49th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024, ICASSP 2024 ; Conference date: 14-04-2024 Through 19-04-2024",

year = "2024",

month = apr,

doi = "10.1109/ICASSP48485.2024.10448120",

language = "English",

isbn = "9798350344868",

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

publisher = "IEEE",

pages = "12931--12935",

booktitle = "2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - Proceedings",

address = "United States",

url = "https://2024.ieeeicassp.org/, https://2024.ieeeicassp.org/program-schedule/, https://ieeexplore.ieee.org/xpl/conhome/10445798/proceeding",

}

Fu, X, Zhang, XL, Yang, CHH & Qi, J 2024, Exploiting A Quantum Multiple Kernel Learning Approach For Low-Resource Spoken Command Recognition. in 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - Proceedings. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, IEEE, pp. 12931-12935, 2024 49th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024, Seoul, Korea, Republic of, 14/04/24. https://doi.org/10.1109/ICASSP48485.2024.10448120

Exploiting A Quantum Multiple Kernel Learning Approach For Low-Resource Spoken Command Recognition. / Fu, Xianyan; Zhang, Xiao Lei; Yang, Chao Han Huck et al.
2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - Proceedings. IEEE, 2024. p. 12931-12935 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Exploiting A Quantum Multiple Kernel Learning Approach For Low-Resource Spoken Command Recognition

AU - Fu, Xianyan

AU - Zhang, Xiao Lei

AU - Yang, Chao Han Huck

AU - Qi, Jun

N1 - This work was supported in part by the National Science Foundation of China (NSFC) under Grant 62176211, and in part by the Project of the Science, Technology, and Innovation Commission of Shenzhen Municipality, China under Grant JCYJ20210324143006016 and JSGG20210802152546026. Publisher Copyright: © 2024 IEEE.

PY - 2024/4

Y1 - 2024/4

N2 - We propose a theoretical analysis of quantum projection learning (QPL) that employs multiple kernels, highlighting its advantages through representation error analysis. Building upon previous studies that utilized a single quantum kernel-based method, we further investigate a quantum projection framework that incorporates multiple Gaussian kernels for low-resource spoken command recognition. Our empirical results align with our theoretical insights, suggesting that methods based on multiple kernels can further enhance the performance of QPL. By leveraging the quantum-to-classical projected output embeddings, we integrate this with a prototypical network for acoustic modeling. When evaluated using Arabic, Chuvash, Irish, and Lithuanian low-resource speech from CommonVoice, our proposed method surpasses the recurrent neural network and single kernel-based classifier baselines by an average of +5.28%.

AB - We propose a theoretical analysis of quantum projection learning (QPL) that employs multiple kernels, highlighting its advantages through representation error analysis. Building upon previous studies that utilized a single quantum kernel-based method, we further investigate a quantum projection framework that incorporates multiple Gaussian kernels for low-resource spoken command recognition. Our empirical results align with our theoretical insights, suggesting that methods based on multiple kernels can further enhance the performance of QPL. By leveraging the quantum-to-classical projected output embeddings, we integrate this with a prototypical network for acoustic modeling. When evaluated using Arabic, Chuvash, Irish, and Lithuanian low-resource speech from CommonVoice, our proposed method surpasses the recurrent neural network and single kernel-based classifier baselines by an average of +5.28%.

KW - low-resource speech classification

KW - multiple kernel learning

KW - quantum kernel projection

UR - http://www.scopus.com/inward/record.url?scp=85195366146&partnerID=8YFLogxK

U2 - 10.1109/ICASSP48485.2024.10448120

DO - 10.1109/ICASSP48485.2024.10448120

M3 - Conference proceeding

AN - SCOPUS:85195366146

SN - 9798350344868

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 12931

EP - 12935

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

PB - IEEE

T2 - 2024 49th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024

Y2 - 14 April 2024 through 19 April 2024

ER -

Fu X, Zhang XL, Yang CHH, Qi J. Exploiting A Quantum Multiple Kernel Learning Approach For Low-Resource Spoken Command Recognition. In 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - Proceedings. IEEE. 2024. p. 12931-12935. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP48485.2024.10448120

Exploiting A Quantum Multiple Kernel Learning Approach For Low-Resource Spoken Command Recognition

Abstract

Publication series

Conference

Scopus Subject Areas

User-Defined Keywords

Access to Document

Other files and links

Cite this