Voice Separation with An Unknown Number of Multipl

专利名称：Voice Separation with An Unknown Number

of Multiple Speakers

发明人：Eliya Nachmani,Lior Wolf,Yossef Mordechay

Adi

申请号：US16853320申请日：20200420

公开号：US20210256993A1公开日：20210819

专利附图：

摘要：In one embodiment, a method includes receiving a mixed audio signal

comprising a mixture of voice signals associated with a plurality of speakers, generating

first audio signals by processing the mixed audio signal using a first machine-learningmodel configured with a first number of output channels, determining that at least oneof the first number of output channels is silent based on the first audio signals,generating second audio signals by processing the mixed audio signal using a secondmachine-learning model configured with a second number of output channels that isfewer than the first number of output channels, determining that each of the secondnumber of output channels is non-silent based on the second audio signals, and using thesecond machine-learning model to separate additional mixed audio signals associatedwith the plurality of speakers.

申请人：Facebook, Inc.

地址：Menlo Park CA US

国籍：US

更多信息请下载全文后查看