Invention Application
- Patent Title: A NEURAL-NETWORK-BASED APPROACH FOR SPEECH DENOISING STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH
-
Application No.: PCT/JP2021/027243Application Date: 2021-07-20
-
Publication No.: WO2022107393A1Publication Date: 2022-05-27
- Inventor: ZHENG Changxi , XU Ruilin , WU Rundi , VONDRICK Carl , ISHIWAKA Yuko
- Applicant: THE TRUSTEES OF COLUMBIA UNIVERSITY IN THE CITY OF NEW YORK , SOFTBANK CORP.
- Applicant Address: 116 Street And Broadway, New York, New York; 1-7-1, Kaigan, Minato-ku, Tokyo
- Assignee: THE TRUSTEES OF COLUMBIA UNIVERSITY IN THE CITY OF NEW YORK,SOFTBANK CORP.
- Current Assignee: THE TRUSTEES OF COLUMBIA UNIVERSITY IN THE CITY OF NEW YORK,SOFTBANK CORP.
- Current Assignee Address: 116 Street And Broadway, New York, New York; 1-7-1, Kaigan, Minato-ku, Tokyo
- Agency: RYUKA IP LAW FIRM
- Priority: US63/116,400 2020-11-20
- Main IPC: G10L21/0208
- IPC: G10L21/0208 ; G10L21/0216 ; G10L21/0232 ; G10L21/0308 ; G10L15/04 ; G10L15/16 ; G10L15/20
Abstract:
Disclosed are methods, systems, device, and other implementations, including a method that includes receiving an audio signal representation, detecting in the received audio signal representation, using a first learning model, one or more silent intervals with reduced foreground sound levels, determining based on the detected one or more silent intervals an estimated full noise profile corresponding to the audio signal representation, and generating with a second learning model, based on the received audio signal representation and on the determined estimated full noise profile, a resultant audio signal representation with a reduced noise level.
Information query
