公开(公告)号：US20220114620A1

公开(公告)日：2022-04-14

申请号：US17518109

申请日：2021-11-03

Applicant: Spotify AB

Inventor： Lu Han ,  Rachel M. Bittner

IPC: G06Q30/02 ,  G10L19/00 ,  G10L21/013 ,  G10L21/0232

Abstract: A call to action processor receives an entity datapoint containing data related to an entity, a campaign objective datapoint containing data associated with a campaign objective, at least one definite script element based on the campaign objective, and entity metadata containing data associated with the entity. The call to action further performs generating at least one variable script element based on the entity metadata, presenting to a device the at least one definite script element the at least one variable script element.

公开(公告)号：US11568256B2

公开(公告)日：2023-01-31

申请号：US17205296

申请日：2021-03-18

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 ,  G10L21/0272 ,  G10L21/028

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：US20200042879A1

公开(公告)日：2020-02-06

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: G06N3/08 ,  G06K9/62 ,  G10H1/00 ,  G06N3/04

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：US20230125789A1

公开(公告)日：2023-04-27

申请号：US18088301

申请日：2022-12-23

Applicant: Spotify AB

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

IPC: G10H1/36 ,  G06N3/082 ,  G06N3/04 ,  G10H1/00 ,  G10L21/0272 ,  G10L21/028 ,  G06F18/214

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：US11195211B2

公开(公告)日：2021-12-07

申请号：US16414387

申请日：2019-05-16

Applicant: Spotify AB

Inventor： Lu Han ,  Rachel M. Bittner

IPC: G06Q30/02 ,  G10L19/00 ,  G10L21/013 ,  G10L21/0232 ,  G06F3/16

Abstract: A call to action processor receives an entity datapoint containing data related to an entity, a campaign objective datapoint containing data associated with a campaign objective, at least one definite script element based on the campaign objective, and entity metadata containing data associated with the entity. The call to action further performs generating at least one variable script element based on the entity metadata, presenting to a device the at least one definite script element the at least one variable script element.

公开(公告)号：US20190355372A1

公开(公告)日：2019-11-21

申请号：US16414381

申请日：2019-05-16

Applicant: Spotify AB

Inventor： Rachel M. Bittner

IPC: G10L21/013 ,  G10L19/00 ,  G10L21/0232

Abstract: Voiceover mixing is provided by receiving a voiceover file and a music file. The voiceover file is audio processed to generate a processed voiceover file and a music file is audio processed to generate a processed music file. The processed voiceover file and the processed music file are weight summed to generate a weighted combination of the processed voiceover file and the processed music file. Single band compressing is performed on the weighted combination. A creative file that contains a compressed and weighted combination of the processed voiceover file and the processed music file is then generated.

公开(公告)号：US20190355024A1

公开(公告)日：2019-11-21

申请号：US16414387

申请日：2019-05-16

Applicant: Spotify AB

Inventor： Lu Han ,  Rachel M. Bittner

IPC: G06Q30/02

Abstract: A call to action processor receives an entity datapoint containing data related to an entity, a campaign objective datapoint containing data associated with a campaign objective, at least one definite script element based on the campaign objective, and entity metadata containing data associated with the entity. The call to action further performs generating at least one variable script element based on the entity metadata, presenting to a device the at least one definite script element the at least one variable script element.

公开(公告)号：US12175957B2

公开(公告)日：2024-12-24

申请号：US18088301

申请日：2022-12-23

Applicant: Spotify AB

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

IPC: G06F15/00 ,  G06F18/214 ,  G06N3/04 ,  G06N3/0464 ,  G06N3/082 ,  G10H1/00 ,  G10H1/36 ,  G10L21/0272 ,  G10L21/028 ,  G10L25/00

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：US11978092B2

公开(公告)日：2024-05-07

申请号：US17518109

申请日：2021-11-03

Applicant: Spotify AB

Inventor： Lu Han ,  Rachel M. Bittner

IPC: G06Q30/0241 ,  G10L19/00 ,  G10L21/013 ,  G10L21/0232 ,  G06F3/16

CPC classification number: G06Q30/0276 ,  G10L19/00 ,  G10L21/013 ,  G10L21/0232 ,  G06F3/165 ,  G10L2021/0135

Abstract: A call to action processor receives an entity datapoint containing data related to an entity, a campaign objective datapoint containing data associated with a campaign objective, at least one definite script element based on the campaign objective, and entity metadata containing data associated with the entity. The call to action further performs generating at least one variable script element based on the entity metadata, presenting to a device the at least one definite script element the at least one variable script element.

公开(公告)号：US20210279588A1

公开(公告)日：2021-09-09

申请号：US17205296

申请日：2021-03-18

Applicant: Spotify AB

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

IPC: 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.