| Project/Area Number |
15K06116
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Measurement engineering
|
|---|
| Research Institution | Prefectural University of Hiroshima |
|---|
Principal Investigator |
IKUTA Akira 県立広島大学, 経営情報学部, 名誉教授 (30145164)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
肖 業貴 県立広島大学, 経営情報学部, 教授 (50252325)
折本 寿子 (益池寿子) 県立広島大学, 経営情報学部, 准教授 (80533207)
|
|---|
| Project Period (FY) |
2015-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2017: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2016: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Fiscal Year 2015: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | 音声信号 / 雑音抑制 / 骨導音 / 気導音 / ベイズフィルタ / 騒音抑制 |
|---|
| Outline of Final Research Achievements |
In this study, in order to suppress the noises that inevitably exist in the observed speech signal of actual environment, a signal processing method to remove the noise for actual speech signals was proposed by jointly using the measured data of bone- and air-conducted speech signals. According to the additive property of sound pressure, the observation of air-conduced speech signal can be expressed as a linear model of the speech signal and a background noise. Furthermore, a propagation model with unknown parameters was introduced as the bone-conducted speech signal model. By introducing Bayes’ theorem based on an orthogonal expansion expression of probability distribution, an algorithm for noise suppression was proposed. By applying the proposed algorithm to real speech signals measured in an anechoic chamber, it was revealed by experiments that better estimation results were obtained by the proposed algorithm as compared with the method based on only air-conducted observations.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究では，音声信号と骨導音の相関情報を活用したベイズフィルタにより音声信号を抽出した。ベイズ推定では，得られる情報（データ）の増加に従い推定精度が向上することから，骨導音と気導音の同時データを測定・活用することにより，それぞれの欠損情報を補うことができる。
音声認識システムは機器点検作業現場や魚・農産物のセリ市場などに導入され，業務効率化に貢献している。また，音声認証により開錠できる生体認証システムも実用化されており，新しいセキュリティとして注目されている。これらの技術で障害となるのは，周囲環境における騒音の存在である。本研究を活用した音声抽出技術の開発により，応用分野の拡大が期待できる。
|
