| 发明名称 |
VIDEO-BASED SOUND SOURCE SEPARATION |
| 摘要 |
A sound source separation method comprising the steps of determining at least one location of at least one sound source based on video data, determining initial estimates of at least two parameters characterizing an audio signal emitted by said sound source, said initial estimates being determined based on said at least one location, performing an expectation-maximization method for determining final estimates of said parameters, and separating the audio signal from a combination of audio signals based on said final estimates of said parameters. |
| 申请公布号 |
US2016180865(A1) |
申请公布日期 |
2016.06.23 |
| 申请号 |
US201514970411 |
申请日期 |
2015.12.15 |
| 申请人 |
CANON KABUSHIKI KAISHA |
发明人 |
Citerin Johann;Kergourlay Gérald |
| 分类号 |
G10L25/57;G06K9/00;G10L21/028 |
主分类号 |
G10L25/57 |
| 代理机构 |
|
代理人 |
|
| 主权项 |
1. A sound source separation method comprising the following steps:
determining at least one location of at least one sound source based on video data, determining at least one parameter characterizing an audio signal emitted by the sound source, the at least one parameter being determined based on the at least one location, and separating the audio signal from a combination of audio signals based on the at least one parameter. |
| 地址 |
Tokyo JP |
