公开(公告)号：US20190104357A1

公开(公告)日：2019-04-04

申请号：US15721644

申请日：2017-09-29

Applicant: Apple Inc.

Inventor： Joshua D. Atkins ,  Mehrez Souden ,  Symeon Delikaris-Manias ,  Peter Raffensperger

IPC: H04R1/40 ,  G06N3/08 ,  G06F17/30

CPC classification number: H04R1/406 ,  G06F16/61 ,  G06N3/08 ,  H04R2201/405 ,  H04R2430/23 ,  H04R2499/11 ,  H04S2400/01 ,  H04S2400/15

Abstract: Impulse responses of a device are measured. A database of sound files is generated by convolving source signals with the impulse responses of the device. The sound files from the database are transformed into time-frequency domain. One or more sub-band directional features is estimated at each sub-band of the time-frequency domain. A deep neural network (DNN) is trained for each sub-band based on the estimated one or more sub-band directional features and a target directional feature.

公开(公告)号：US10334357B2

公开(公告)日：2019-06-25

申请号：US15721644

申请日：2017-09-29

Applicant: Apple Inc.

Inventor： Joshua D. Atkins ,  Mehrez Souden ,  Symeon Delikaris-Manias ,  Peter Raffensperger

IPC: H04R1/40 ,  G06N3/08 ,  G06F16/61

Abstract: Impulse responses of a device are measured. A database of sound files is generated by convolving source signals with the impulse responses of the device. The sound files from the database are transformed into time-frequency domain. One or more sub-band directional features is estimated at each sub-band of the time-frequency domain. A deep neural network (DNN) is trained for each sub-band based on the estimated one or more sub-band directional features and a target directional feature.