Curriculum Vitaes
Profile Information
- Affiliation
- Professor, Faculty of Science and Technology Department of Science and Technology , Seikei University
- Degree
- 博士(工学)(慶應義塾大学)
- J-GLOBAL ID
- 201501025877783683
- researchmap Member ID
- B000244685
Research Interests
2Research Areas
1Research History
5-
Apr, 2015 - Mar, 2017
-
Jan, 2009 - Jul, 2009
-
Apr, 1996 - Jul, 1998
Education
3-
Apr, 1994 - Mar, 1996
-
Apr, 1990 - Mar, 1994
Committee Memberships
2-
Nov, 2022 - Present
-
Nov, 2022 - Mar, 2023
Awards
11-
Aug, 2017
-
Jun, 2012
Papers
25-
2021 IEEE International Conference on Consumer Electronics (ICCE), Jan 10, 2021 Peer-reviewed
-
Artificial Intelligence: Concepts, Methodologies, Tools, and Applications, 3 2021-2037, Dec 12, 2016 Peer-reviewedTied-mixture HMMs have been proposed as the acoustic model for large-vocabulary continuous speech recognition and have yielded promising results. They share base-distribution and provide more flexibility in choosing the degree of tying than state-clustered HMMs. However, it is unclear which acoustic models to superior to the other under the same training data. Moreover, LBG algorithm and EM algorithm, which are the usual training methods for HMMs, have not been compared. Therefore in this paper, the recognition performance of the respective HMMs and the respective training methods are compared under the same condition. It was found that the number of parameters and the word error rate for both HMMs are equivalent when the number of codebooks is sufficiently large. It was also found that training method using the LBG algorithm achieves a 90% reduction in training time compared to training method using the EM algorithm, without degradation of recognition accuracy.
-
INTERNATIONAL JOURNAL OF MULTIMEDIA DATA ENGINEERING & MANAGEMENT, 7(2) 53-67, Apr, 2016 Peer-reviewed
-
The Journal of the Faculty of Science and Technology, Seikei University, 52(2) 5-10, Dec, 2015
-
Journal of Information Technology Research, 7(3) 15-31, Jul 1, 2014 Peer-reviewedTied-mixture HMMs have been proposed as the acoustic model for large-vocabulary continuous speech recognition and have yielded promising results. They share base-distribution and provide more flexibility in choosing the degree of tying than state-clustered HMMs. However, it is unclear which acoustic models to superior to the other under the same training data. Moreover, LBG algorithm and EM algorithm, which are the usual training methods for HMMs, have not been compared. Therefore in this paper, the recognition performance of the respective HMMs and the respective training methods are compared under the same condition. It was found that the number of parameters and the word error rate for both HMMs are equivalent when the number of codebooks is sufficiently large. It was also found that training method using the LBG algorithm achieves a 90% reduction in training time compared to training method using the EM algorithm, without degradation of recognition accuracy.
Misc.
29-
50(2) 575-586, Feb 15, 2009
-
IEICE technical report, 107(551) 123-128, Mar 13, 2008It could be useful to apply other speaker's speech data to a concatenative speech synthesis system for complementing lack of original speaker's speech database. As a basic study to investigate ability of phoneme replacement, vowels or long vowels of /CVC/ triphone environment in a continuous speech were replaced by other 24 speakers' ones and perceptual naturalness was evaluated. As a result, it tends to show higher naturalness at 1) specified 3 speakers, 2) length of replaced phoneme is short or 3) F0 of replaced phoneme is lower. Furthermore correlation analysis between evaluated value and spectral feature suggests that 1) there is higher correlation at specified 3 frequency bands and 2) spectral peak or dip is related with evaluated value.
-
Proceedings of the Spoken Document Processing Workshop, 2 51-58, Feb 29, 2008
-
Segmental Selection from Broadcast News Recordings for a High Quality Concatenative Speech Synthesis33(4) 209-214, Jun 26, 2003
-
Segmental Selection from Broadcast News Recridings for a High Quality Concatenative Speech SynthesisIEICE technical report. Speech, 103(154) 1-6, Jun 26, 2003Conventional speech databases of the concatenative speech synthesis systems are prepared by using the sentences read with appropriate speed and reading style for speech synthesis. If we can use broadcast programs in our archives as a speech database, various sentences can be synthesized naturally. So in this paper, we try to use Japanese broadcast news programs as a speech database and use words and triphones as search units in order to search the huge speech database in shorter time. We got 71% preference score in the paired comparison test and 2.6 mean opinion score in five-point MOS test.
-
IEICE transactions on information and systems, 85(2) 436-436, Feb 1, 2002This paper proposes a method of constructing acoustic models from training data clustered in two stages. The first stage generates cluster models from small-scale training data gathered form a target task. The second stage clusters a large-scale database based on the cluster models. In decoding, the best acoustic model is selected from all the acoustic models based on the GMM likelihood using some initial frames of an input utterance. Broadcast news transcription experiments showed that the proposed models achieved a word error reduction of 20% and a processing time reduction of 22%, compared with a non-clustered model.
-
IPSJ SIG Notes, 2001(123) 1-6, Dec 20, 2001A new method for estimating dynamic features used in speech recognition is proposed. The proposed method involves: 1)increasing time resolution of static features used for dynamic feature estimation, 2)relaxing the restriction that estimation span must be an integer multiple of frame shift, and making the estimation span arbitrary. Evaluating the proposed method with a broadcast news testset, word accuracy improved by selecting appropriate estimation span. Error reduction rate in word accuracy was up to 24% relatively in noisy news speech. The overall error reduction rate was 6.5% relatively.
-
IEICE technical report. Natural language understanding and models of communication, 101(520) 1-6, Dec 13, 2001A new method for estimating dynamic features used in speech recognition is proposed. The proposed method involves:1)increasing time resolution of static features used for dynamic feature estimation, 2)relaxing the restriction that estimation span must be an integer multiple of frame shift, and making the estimation span arbitrary. Evaluating the proposed method with a broadcast news testset, word accuracy improved by selecting appropriate estimation span. Error reduction rate in word accuracy was up to 24% relatively in noisy news speech. The overall error reduction rate was 6.5% relatively.
-
2001(2) 179-180, Oct 1, 2001
-
2001(2) 139-140, Oct 1, 2001
-
2001(2) 37-38, Oct 1, 2001
-
IEICE technical report. Speech, 101(156) 1-6, Jun 22, 2001We are studying speech recognition for simultaneous subtitling of news programs. One of our current studies is recognition of reporter's speech in the afflicted area, from overseas and so on. Because these speeches are uttered in various acoustic conditions, we need technologies to handle such conditions. In this paper, a filter-bank subtraction technique is proposed for robust speech recognition under various acoustic conditions. It calculates, in each pass band of the filter-bank, the minimum value of the filter output in a certain interval of time, and then estimates a noise spectrum over all bands as a composition of those minimum values. The experiments show that the new method yielded better recognition performance than the conventional spectral subtraction techniques.
-
2001(1) 59-60, Mar 1, 2001
-
2000(2) 115-116, Sep 1, 2000
-
IEICE technical report. Speech, 100(136) 9-15, Jun 15, 2000This paper proposes a method of constructing acoustic models for automatic news subtitling. In this method, two-stage clustering is used for an adaptation-data selection from a large-scale database. The first stage generates GMMs from a small amount of data and second atage clusters a large-scale database by the GMMs. A broadcast news transcription experiment showed that the proposed method reduced in toth a word error rate and processing time. The best result in the experiment showed 33% WER reduction and 25% decoding time reduction compared with the baseline method, which had no clustering. Furthermore, we found that it takes just 0.4 seconds from the beginning of an input utterance to select an HMM correctly. The result showed 18% processing time reduction in total compared with the baseline method.
-
IEICE technical report. Speech, 99(393) 31-36, Oct 28, 1999In this paper, the effects of the various acoustic conditions on the speech recognition are reported. A broadcast transcription system for simultaneous subtitling of news program is being developed. Speech recognition for studio announcers' speech, especially for announcers' read speech, has been investigated so far. To realize simultaneous subtitling for the whole news program, speech in various acoustic conditions should be recognized. For example, speech spoken by people other than announcers or speech with background sound such as noise and music become the recognition object. Speech in the whole news program was recognized to investigate the effort of the acoustic condition to the recognition accuracy.
Books and Other Publications
1Presentations
47Teaching Experience
8-
応用プログラミング(2年前期再履修者向けC++) (成蹊大学)
-
情報科学特別実験(大学院) (成蹊大学)
-
情報科学プロジェクト実験(3年ゼミ) (成蹊大学)
-
パターン認識 (成蹊大学)
-
フレッシャーズセミナー(1年前期ゼミ) (成蹊大学)
Professional Memberships
4Research Projects
1-
平成28年度助成, 公益財団法人放送文化基金, Apr, 2017 - Mar, 2018
お知らせ
【連絡事項】
成蹊大学専任教員各位
Ufinityにログイン後、「編集」ボタンを押した際に「権限が不正」というエラーメッセージが生じた際は、総合企画課までご連絡ください。
総合企画課：kikaku@jim.seikei.ac.jp
※各教員の担当科目については、以下のリンク先にある「教員名」に該当教員の氏名を入力の上、検索してください。