-
公开(公告)号：US20230343348A1
公开(公告)日：2023-10-26
申请号：US18344567
申请日：2023-06-29
Applicant: Google LLC
Inventor： Jesse Engel , Adam Roberts , Chenjie Gu , Lamtharn Hantrakul
Abstract: Systems and methods of the present disclosure are directed toward digital signal processing using machine-learned differentiable digital signal processors. For example, embodiments of the present disclosure may include differentiable digital signal processors within the training loop of a machine-learned model (e.g., for gradient-based training). Advantageously, systems and methods of the present disclosure provide high quality signal processing using smaller models than prior systems, thereby reducing energy costs (e.g., storage and/or processing costs) associated with performing digital signal processing.
-
公开(公告)号：US20240233713A1
公开(公告)日：2024-07-11
申请号：US18412394
申请日：2024-01-12
Applicant: Google LLC
Inventor： Andrea Agostinelli , Timo Immanuel Denk , Antoine Caillon , Neil Zeghidour , Jesse Engel , Mauro Verzetti , Christian Frank , Zalán Borsos , Matthew Sharifi , Adam Joseph Roberts , Marco Tagliasacchi
IPC: G10L15/16 , G06N3/0455 , G06N3/0475 , G10H1/00 , G10L15/06 , G10L15/18
CPC classification number: G10L15/16 , G06N3/0455 , G06N3/0475 , G10H1/0008 , G10L15/063 , G10L15/1815 , G10H2210/056 , G10H2250/311
Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a prediction of an audio signal. One of the methods includes receiving a request to generate an audio signal conditioned on an input; processing the input using an embedding neural network to map the input to one or more embedding tokens; generating a semantic representation of the audio signal; generating, using one or more generative neural networks and conditioned on at least the semantic representation and the embedding tokens, an acoustic representation of the audio signal; and processing at least the acoustic representation using a decoder neural network to generate the prediction of the audio signal.
-
公开(公告)号：US11915689B1
公开(公告)日：2024-02-27
申请号：US18463196
申请日：2023-09-07
Applicant: Google LLC
Inventor： Andrea Agostinelli , Timo Immanuel Denk , Antoine Caillon , Neil Zeghidour , Jesse Engel , Mauro Verzetti , Christian Frank , Zalán Borsos , Matthew Sharifi , Adam Joseph Roberts , Marco Tagliasacchi
CPC classification number: G10L15/16 , G10H1/0008 , G10L15/063 , G10L15/1815 , G10H2210/056 , G10H2250/311
Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a prediction of an audio signal. One of the methods includes receiving a request to generate an audio signal conditioned on an input; processing the input using an embedding neural network to map the input to one or more embedding tokens; generating a semantic representation of the audio signal; generating, using one or more generative neural networks and conditioned on at least the semantic representation and the embedding tokens, an acoustic representation of the audio signal; and processing at least the acoustic representation using a decoder neural network to generate the prediction of the audio signal.
-
Patent Agency Ranking
公开(公告)号：US12080311B2
公开(公告)日：2024-09-03
申请号：US18344567
申请日：2023-06-29
Applicant: Google LLC
Inventor： Jesse Engel , Adam Roberts , Chenjie Gu , Lamtharn Hantrakul
Abstract: Systems and methods of the present disclosure are directed toward digital signal processing using machine-learned differentiable digital signal processors. For example, embodiments of the present disclosure may include differentiable digital signal processors within the training loop of a machine-learned model (e.g., for gradient-based training). Advantageously, systems and methods of the present disclosure provide high quality signal processing using smaller models than prior systems, thereby reducing energy costs (e.g., storage and/or processing costs) associated with performing digital signal processing.
-
公开(公告)号：US20240079001A1
公开(公告)日：2024-03-07
申请号：US18463196
申请日：2023-09-07
Applicant: Google LLC
Inventor： Andrea Agostinelli , Timo Immanuel Denk , Antoine Caillon , Neil Zeghidour , Jesse Engel , Mauro Verzetti , Christian Frank , Zalán Borsos , Matthew Sharifi , Adam Joseph Roberts
CPC classification number: G10L15/16 , G10H1/0008 , G10L15/063 , G10L15/1815 , G10H2210/056 , G10H2250/311
Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a prediction of an audio signal. One of the methods includes receiving a request to generate an audio signal conditioned on an input; processing the input using an embedding neural network to map the input to one or more embedding tokens; generating a semantic representation of the audio signal; generating, using one or more generative neural networks and conditioned on at least the semantic representation and the embedding tokens, an acoustic representation of the audio signal; and processing at least the acoustic representation using a decoder neural network to generate the prediction of the audio signal.
-
公开(公告)号：US20240395277A1
公开(公告)日：2024-11-28
申请号：US18792298
申请日：2024-08-01
Applicant: Google LLC
Inventor： Jesse Engel , Adam Roberts , Chenjie Gu , Lamtharn Hantrakul
Abstract: Systems and methods of the present disclosure are directed toward digital signal processing using machine-learned differentiable digital signal processors. For example, embodiments of the present disclosure may include differentiable digital signal processors within the training loop of a machine-learned model (e.g., for gradient-based training). Advantageously, systems and methods of the present disclosure provide high quality signal processing using smaller models than prior systems, thereby reducing energy costs (e.g., storage and/or processing costs) associated with performing digital signal processing.
-
公开(公告)号：US11735197B2
公开(公告)日：2023-08-22
申请号：US16922543
申请日：2020-07-07
Applicant: Google LLC
Inventor： Jesse Engel , Adam Roberts , Chenjie Gu , Lamtharn Hantrakul
Abstract: Systems and methods of the present disclosure are directed toward digital signal processing using machine-learned differentiable digital signal processors. For example, embodiments of the present disclosure may include differentiable digital signal processors within the training loop of a machine-learned model (e.g., for gradient-based training). Advantageously, systems and methods of the present disclosure provide high quality signal processing using smaller models than prior systems, thereby reducing energy costs (e.g., storage and/or processing costs) associated with performing digital signal processing.
-
公开(公告)号：US20220013132A1
公开(公告)日：2022-01-13
申请号：US16922543
申请日：2020-07-07
Applicant: Google LLC
Inventor： Jesse Engel , Adam Roberts , Chenjie Gu , Lamtharn Hantrakul
Abstract: Systems and methods of the present disclosure are directed toward digital signal processing using machine-learned differentiable digital signal processors. For example, embodiments of the present disclosure may include differentiable digital signal processors within the training loop of a machine-learned model (e.g., for gradient-based training). Advantageously, systems and methods of the present disclosure provide high quality signal processing using smaller models than prior systems, thereby reducing energy costs (e.g., storage and/or processing costs) associated with performing digital signal processing.
-
-
-
-
-
-
-
Search Results
8
records
(took 0.012 seconds)
