Study of Formant Modification for Children ASR

Hemant Kathania; Sudarsana Kadiri; Paavo Alku; Mikko Kurimo

doi:10.1109/ICASSP40776.2020.9053334

Study of Formant Modification for Children ASR

Hemant Kathania, Sudarsana Kadiri, Paavo Alku, Mikko Kurimo

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

38 Citations (Scopus)

248 Downloads (Pure)

Abstract

The performance of automatic speech recognition systems for children’s speech is known to suffer from the large variation and mismatch in the acoustic and linguistic attributes between children’s and adults’ speech. One of the various identified sources of mismatch is the difference in formant frequencies between adults and children. In this paper, we propose a formant modification method to mitigate differences between adults’ and children’s speech and to improve the performance of ASR for children. The explored technique gives a relative 27% improvement in system performance compared to a hybrid DNN-HMM baseline. We also compare the system performance with related speaker adaptation methods like vocal tract length normalization (VTLN) and speaking rate adapta-
tion (SRA) and find that the proposed method gives improvements over them, as well. Combining the proposed method with VTLN and SRA results in a further reduction of WER. We also found that the proposed method performs well even
for noisy speech.

Original language	English
Title of host publication	2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Proceedings
Publisher	IEEE
Pages	7429-7433
Number of pages	5
ISBN (Electronic)	978-1-5090-6631-5
ISBN (Print)	978-1-5090-6632-2
DOIs	https://doi.org/10.1109/ICASSP40776.2020.9053334
Publication status	Published - May 2020
MoE publication type	A4 Conference publication
Event	IEEE International Conference on Acoustics, Speech, and Signal Processing - Virtual conference, Barcelona, Spain Duration: 4 May 2020 → 8 May 2020 Conference number: 45

Publication series

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

Conference

Conference	IEEE International Conference on Acoustics, Speech, and Signal Processing
Abbreviated title	ICASSP
Country/Territory	Spain
City	Barcelona
Period	04/05/2020 → 08/05/2020
Other	Virtual conference

Keywords

hildren speech recognition
Formant modification
DNN

Access to Document

10.1109/ICASSP40776.2020.9053334

Children_ASR_ICASSP_2020Accepted author manuscript, 219 KB

Cite this

Kathania, H., Kadiri, S., Alku, P., & Kurimo, M. (2020). Study of Formant Modification for Children ASR. In 2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Proceedings (pp. 7429-7433). Article 9053334 (Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing). IEEE. https://doi.org/10.1109/ICASSP40776.2020.9053334

@inproceedings{c2ad5b2ee1444baf8a7d69c27a029724,

title = "Study of Formant Modification for Children ASR",

abstract = "The performance of automatic speech recognition systems for children{\textquoteright}s speech is known to suffer from the large variation and mismatch in the acoustic and linguistic attributes between children{\textquoteright}s and adults{\textquoteright} speech. One of the various identified sources of mismatch is the difference in formant frequencies between adults and children. In this paper, we propose a formant modification method to mitigate differences between adults{\textquoteright} and children{\textquoteright}s speech and to improve the performance of ASR for children. The explored technique gives a relative 27% improvement in system performance compared to a hybrid DNN-HMM baseline. We also compare the system performance with related speaker adaptation methods like vocal tract length normalization (VTLN) and speaking rate adapta-tion (SRA) and find that the proposed method gives improvements over them, as well. Combining the proposed method with VTLN and SRA results in a further reduction of WER. We also found that the proposed method performs well evenfor noisy speech.",

keywords = "hildren speech recognition, Formant modification, DNN",

author = "Hemant Kathania and Sudarsana Kadiri and Paavo Alku and Mikko Kurimo",

year = "2020",

month = may,

doi = "10.1109/ICASSP40776.2020.9053334",

language = "English",

isbn = "978-1-5090-6632-2",

series = "Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing",

publisher = "IEEE",

pages = "7429--7433",

booktitle = "2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Proceedings",

address = "United States",

note = "IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP ; Conference date: 04-05-2020 Through 08-05-2020",

}

Kathania, H, Kadiri, S, Alku, P & Kurimo, M 2020, Study of Formant Modification for Children ASR. in 2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Proceedings., 9053334, Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing, IEEE, pp. 7429-7433, IEEE International Conference on Acoustics, Speech, and Signal Processing, Barcelona, Spain, 04/05/2020. https://doi.org/10.1109/ICASSP40776.2020.9053334

Study of Formant Modification for Children ASR. / Kathania, Hemant; Kadiri, Sudarsana; Alku, Paavo et al.
2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Proceedings. IEEE, 2020. p. 7429-7433 9053334 (Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Study of Formant Modification for Children ASR

AU - Kathania, Hemant

AU - Kadiri, Sudarsana

AU - Alku, Paavo

AU - Kurimo, Mikko

N1 - Conference code: 45

PY - 2020/5

Y1 - 2020/5

N2 - The performance of automatic speech recognition systems for children’s speech is known to suffer from the large variation and mismatch in the acoustic and linguistic attributes between children’s and adults’ speech. One of the various identified sources of mismatch is the difference in formant frequencies between adults and children. In this paper, we propose a formant modification method to mitigate differences between adults’ and children’s speech and to improve the performance of ASR for children. The explored technique gives a relative 27% improvement in system performance compared to a hybrid DNN-HMM baseline. We also compare the system performance with related speaker adaptation methods like vocal tract length normalization (VTLN) and speaking rate adapta-tion (SRA) and find that the proposed method gives improvements over them, as well. Combining the proposed method with VTLN and SRA results in a further reduction of WER. We also found that the proposed method performs well evenfor noisy speech.

AB - The performance of automatic speech recognition systems for children’s speech is known to suffer from the large variation and mismatch in the acoustic and linguistic attributes between children’s and adults’ speech. One of the various identified sources of mismatch is the difference in formant frequencies between adults and children. In this paper, we propose a formant modification method to mitigate differences between adults’ and children’s speech and to improve the performance of ASR for children. The explored technique gives a relative 27% improvement in system performance compared to a hybrid DNN-HMM baseline. We also compare the system performance with related speaker adaptation methods like vocal tract length normalization (VTLN) and speaking rate adapta-tion (SRA) and find that the proposed method gives improvements over them, as well. Combining the proposed method with VTLN and SRA results in a further reduction of WER. We also found that the proposed method performs well evenfor noisy speech.

KW - hildren speech recognition

KW - Formant modification

KW - DNN

UR - http://www.scopus.com/inward/record.url?scp=85089213866&partnerID=8YFLogxK

U2 - 10.1109/ICASSP40776.2020.9053334

DO - 10.1109/ICASSP40776.2020.9053334

M3 - Conference article in proceedings

SN - 978-1-5090-6632-2

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

SP - 7429

EP - 7433

BT - 2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Proceedings

PB - IEEE

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

Y2 - 4 May 2020 through 8 May 2020

ER -

Study of Formant Modification for Children ASR

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Science-IT

Cite this

Study of Formant Modification for Children ASR

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Equipment

Science-IT

Cite this