公开(公告)号：US10991385B2

公开(公告)日：2021-04-27

申请号：US16165498

申请日：2018-10-19

Applicant: Spotify AB

Inventor： Andreas Simon Thore Jansson ,  Angus William Sackfield ,  Ching Chuan Sung ,  David Rubinstein

IPC: G10L25/81 ,  G10L21/06 ,  G10L25/18 ,  G10L15/16 ,  G06N3/08

Abstract: A system, method and computer product for estimating a component of a provided audio signal. The method comprises converting the provided audio signal to an image, processing the image with a neural network trained to estimate one of vocal content and instrumental content, and storing a spectral mask output from the neural network as a result of the image being processed by the neural network. The neural network is a U-Net. The method also comprises providing the spectral mask to a client media playback device, which applies the spectral mask to a spectrogram of the provided audio signal, to provide a masked spectrogram. The media playback device also transforms the masked spectrogram to an audio signal, and plays back that audio signal via an output user interface.

公开(公告)号：US10977555B2

公开(公告)日：2021-04-13

申请号：US16521756

申请日：2019-07-25

Applicant: Spotify AB

Inventor： Andreas Simon Thore Jansson ,  Angus William Sackfield ,  Ching Chuan Sung ,  Rachel M. Bittner

IPC: G06F15/00 ,  G10L25/00 ,  G06N3/08 ,  G06K9/62 ,  G06N3/04 ,  G10H1/00

Abstract: A system, method and computer product for training a neural network system. The method comprises inputting an audio signal to the system to generate plural outputs f(X, Θ). The audio signal includes one or more of vocal content and/or musical instrument content, and each output f(X, Θ) corresponds to a respective one of the different content types. The method also comprises comparing individual outputs f(X, Θ) of the neural network system to corresponding target signals. For each compared output f(X, Θ), at least one parameter of the system is adjusted to reduce a result of the comparing performed for the output f(X, Θ), to train the system to estimate the different content types. In one example embodiment, the system comprises a U-Net architecture. After training, the system can estimate various different types of vocal and/or instrument components of an audio signal, depending on which type of component(s) the system is trained to estimate.

公开(公告)号：US10923142B2

公开(公告)日：2021-02-16

申请号：US16242525

申请日：2019-01-08

Applicant: Spotify AB

Inventor： Andreas Simon Thore Jansson ,  Angus William Sackfield ,  Ching Chuan Sung

IPC: G06F15/00 ,  G10L25/00 ,  G10L25/81 ,  G06N3/08 ,  G06N5/04 ,  G10L15/16 ,  G10L21/10

Abstract: A system, method and computer product for training a neural network system. The method comprises applying an audio signal to the neural network system, the audio signal including a vocal component and a non-vocal component. The method also comprises comparing an output of the neural network system to a target signal, and adjusting at least one parameter of the neural network system to reduce a result of the comparing, for training the neural network system to estimate one of the vocal component and the non-vocal component. In one example embodiment, the system comprises a U-Net architecture. After training, the system can estimate vocal or instrumental components of an audio signal, depending on which type of component the system is trained to estimate.