Information: Join today and make your research activities more affordable! Technical workshop participation fees and annual registration fees are available at member rates.
Notice: [Important] Announcement of Changes to Registration Fee Payment and Manuscript Upload Procedures for IEICE Technical Meetings
IEICE Technical Committee Submission System
Conference Paper's Information		 Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2023-02-28 09:50
End-to-End Speech Synthesis Based on Articulatory Movements Captured by Real-time MRI
Yuto Otani, Shun Sawada, Hidefumi Ohmura, Kouichi Katsurada (Tokyo Univ. Sci.) EA2022-77 SIP2022-121 SP2022-41
Abstract (in Japanese) (See Japanese page) 
(in English) We propose an end-to-end deep learning model for speech synthesis based on articulatory movements captured by real-time MRI (rtMRI). Previous methods for speech synthesis from rtMRI videos only predict parameters of vocal tract shape and require additional information about fundamental frequency. In this work, we propose a model using a CNN-BiLSTM to predict the mel-spectrogram as an intermediate representation, and a HiFi-GAN vocoder to synthesize speech. We evaluated speech quality and the fundamental frequency accuracy of the synthesized speech on the rtMRI dataset containing ATR 503 sentences read by a single speaker. The experimental results show that the PESQ score and the RMSE of F0 are 1.64 and 26.7 Hz, respectively, which demonstrates that the entire acoustic parameters including the fundamental frequency can be estimated from rtMRI videos. In the experiment on USC-TIMIT database, we could get good PESQ score and RMSE of F0. However, the synthesized speech is unclear, which indicates quality of the datasets significantly affects intelligibleness of the synthesized speech.
Keyword (in Japanese) (See Japanese page) 
(in English) real-time MRI / articulatory movement / speech synthesis / end-to-end / / / /  
Reference Info. IEICE Tech. Rep., vol. 122, no. 389, SP2022-41, pp. 13-18, Feb. 2023.
Paper # SP2022-41 
Date of Issue 2023-02-21 (EA, SIP, SP) 
ISSN Online edition: ISSN 2432-6380
Copyright
and
reproduction		 All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034)
Download PDF EA2022-77 SIP2022-121 SP2022-41

Conference Information
Committee SP IPSJ-SLP EA SIP  
Conference Date 2023-02-28 - 2023-03-01 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To SP 
Conference Code 2023-02-SP-SLP-EA-SIP 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) End-to-End Speech Synthesis Based on Articulatory Movements Captured by Real-time MRI 
Sub Title (in English)  
Keyword(1) real-time MRI  
Keyword(2) articulatory movement  
Keyword(3) speech synthesis  
Keyword(4) end-to-end  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Yuto Otani  
1st Author's Affiliation Tokyo University of Science (Tokyo Univ. Sci.)
2nd Author's Name Shun Sawada  
2nd Author's Affiliation Tokyo University of Science (Tokyo Univ. Sci.)
3rd Author's Name Hidefumi Ohmura  
3rd Author's Affiliation Tokyo University of Science (Tokyo Univ. Sci.)
4th Author's Name Kouichi Katsurada  
4th Author's Affiliation Tokyo University of Science (Tokyo Univ. Sci.)
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
21st Author's Name  
21st Author's Affiliation ()
22nd Author's Name  
22nd Author's Affiliation ()
23rd Author's Name  
23rd Author's Affiliation ()
24th Author's Name  
24th Author's Affiliation ()
25th Author's Name  
25th Author's Affiliation ()
26th Author's Name / /
26th Author's Affiliation ()
()
27th Author's Name / /
27th Author's Affiliation ()
()
28th Author's Name / /
28th Author's Affiliation ()
()
29th Author's Name / /
29th Author's Affiliation ()
()
30th Author's Name / /
30th Author's Affiliation ()
()
31st Author's Name / /
31st Author's Affiliation ()
()
32nd Author's Name / /
32nd Author's Affiliation ()
()
33rd Author's Name / /
33rd Author's Affiliation ()
()
34th Author's Name / /
34th Author's Affiliation ()
()
35th Author's Name / /
35th Author's Affiliation ()
()
36th Author's Name / /
36th Author's Affiliation ()
()
Speaker Author-1 
Date Time 2023-02-28 09:50:00 
Presentation Time 20 minutes 
Registration for SP 
Paper # EA2022-77, SIP2022-121, SP2022-41 
Volume (vol) vol.122 
Number (no) no.387(EA), no.388(SIP), no.389(SP) 
Page pp.13-18 
#Pages 6 
Date of Issue 2023-02-21 (EA, SIP, SP) 

[Return to Top Page]

[Return to IEICE Web Page]
The Institute of Electronics, Information and Communication Engineers (IEICE), Japan