公开(公告)号：US20210089909A1

公开(公告)日：2021-03-25

申请号：US17032578

申请日：2020-09-25

Applicant: DeepMind Technologies Limited

Inventor： Mikolaj Binkowski ,  Karen Simonyan ,  Jeffrey Donahue ,  Aidan Clark ,  Sander Etienne Lea Dieleman ,  Erich Konrad Elsen ,  Luis Carlos Cobo Rus ,  Norman Casagrande

IPC: G06N3/08 ,  G06N3/04 ,  G06K9/62

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating output audio examples using a generative neural network. One of the methods includes obtaining a training conditioning text input; processing a training generative input comprising the training conditioning text input using a feedforward generative neural network to generate a training audio output; processing the training audio output using each of a plurality of discriminators, wherein the plurality of discriminators comprises one or more conditional discriminators and one or more unconditional discriminators; determining a first combined prediction by combining the respective predictions of the plurality of discriminators; and determining an update to current values of a plurality of generative parameters of the feedforward generative neural network to increase a first error in the first combined prediction.

公开(公告)号：US12211484B2

公开(公告)日：2025-01-28

申请号：US18418025

申请日：2024-01-19

Applicant: DeepMind Technologies Limited

Inventor： Luis Carlos Cobo Rus ,  Nal Kalchbrenner ,  Erich Elsen ,  Chenjie Gu

IPC: G10L25/30 ,  G10L13/00 ,  G10L13/047 ,  G10L13/08

Abstract: Techniques are disclosed that enable generation of an audio waveform representing synthesized speech based on a difference signal determined using an autoregressive model. Various implementations include using a distribution of the difference signal values to represent sounds found in human speech with a higher level of granularity than sounds not frequently found in human speech. Additional or alternative implementations include using one or more speakers of a client device to render the generated audio waveform.

公开(公告)号：US20220254330A1

公开(公告)日：2022-08-11

申请号：US17610934

申请日：2019-05-20

Applicant: DeepMind Technologies Limited

Inventor： Luis Carlos Cobo Rus ,  Nal Kalchbrenner ,  Erich Elsen ,  Chenjie Gu

IPC: G10L13/047 ,  G10L13/08 ,  G10L25/30

Abstract: Techniques are disclosed that enable generation of an audio waveform representing synthesized speech based on a difference signal determined using an autoregressive model. Various implementations include using a distribution of the difference signal values to represent sounds found in human speech with a higher level of granularity than sounds not frequently found in human speech. Additional or alternative implementations include using one or more speakers of a client device to render the generated audio waveform.

公开(公告)号：US20240161729A1

公开(公告)日：2024-05-16

申请号：US18418025

申请日：2024-01-19

Applicant: DeepMind Technologies Limited

Inventor： Luis Carlos Cobo Rus ,  Nal Kalchbrenner ,  Erich Elsen ,  Chenjie Gu

IPC: G10L13/047 ,  G10L13/08 ,  G10L25/30

CPC classification number: G10L13/047 ,  G10L13/08 ,  G10L25/30

Abstract: Techniques are disclosed that enable generation of an audio waveform representing synthesized speech based on a difference signal determined using an autoregressive model. Various implementations include using a distribution of the difference signal values to represent sounds found in human speech with a higher level of granularity than sounds not frequently found in human speech. Additional or alternative implementations include using one or more speakers of a client device to render the generated audio waveform.

公开(公告)号：US11915682B2

公开(公告)日：2024-02-27

申请号：US17610934

申请日：2019-05-20

Applicant: DeepMind Technologies Limited

Inventor： Luis Carlos Cobo Rus ,  Nal Kalchbrenner ,  Erich Elsen ,  Chenjie Gu

IPC: G10L13/00 ,  G10L19/00 ,  G10L13/047 ,  G10L13/08 ,  G10L25/30

CPC classification number: G10L13/047 ,  G10L13/08 ,  G10L25/30

Abstract: Techniques are disclosed that enable generation of an audio waveform representing synthesized speech based on a difference signal determined using an autoregressive model. Various implementations include using a distribution of the difference signal values to represent sounds found in human speech with a higher level of granularity than sounds not frequently found in human speech. Additional or alternative implementations include using one or more speakers of a client device to render the generated audio waveform.