Title:
信号処理装置、信号処理方法、信号処理プログラム、学習装置、学習方法及び学習プログラム
Document Type and Number:
Japanese Patent JP7293162
Kind Code:
B2
Abstract:
To precisely extract a voice signal of a target speaker.SOLUTION: A first conversion part 111 converts a voice signal of a time domain, which is obtained from utterance of a target speaker, into an adaptation feature amount. A second conversion part 121 converts a mixture voice signal of a multichannel time domain, which is obtained by recording voices of a plurality of sound sources by a plurality of microphones, into a pre-adaptation feature amount by a neural network. A third conversion part 123 converts a post-adaptation feature amount, which is obtained by adapting the pre-adaptation feature amount to the target speaker by using the adaptation feature amount, into information for output by the neural network.SELECTED DRAWING: Figure 1
Inventors:
Delcroix Mark
Tsubasa Ochiai
Keisuke Kinoshita
Naohiro Tawara
Tomohiro Nakatani
Akiko Araki
Moricova Katerina
Tsubasa Ochiai
Keisuke Kinoshita
Naohiro Tawara
Tomohiro Nakatani
Akiko Araki
Moricova Katerina
Application Number:
JP2020070081A
Publication Date:
June 19, 2023
Filing Date:
April 08, 2020
Export Citation:
Assignee:
Nippon Telegraph and Telephone Corporation
Brno University of Technology
Brno University of Technology
International Classes:
G10L17/18; G10L21/0272; G10L21/028; G10L21/0308; G10L25/30
Domestic Patent References:
| JP2020181060A |
Foreign References:
| WO2019198306A1 |
Attorney, Agent or Firm:
Sakai International Patent Office
Previous Patent: Conductive composition, metallized substrate and manufacturing method thereof
Next Patent: GATE FENCE
Next Patent: GATE FENCE