公开(公告)号：US20240296857A1

公开(公告)日：2024-09-05

申请号：US18575421

申请日：2022-07-05

申请人： Pink Nectarine Health AB

发明人： Martin LINREY

IPC分类号： G10L21/034 ,  G10L21/0364 ,  G10L25/84 ,  H04W4/80 ,  H04W40/02

CPC分类号： G10L21/034 ,  G10L21/0364 ,  G10L25/84 ,  H04W4/80 ,  H04W40/02

摘要： Methods, a system, a network of nodes and at least one wearable device, where the at least one wearable device is adapted to be worn by individuals and are configured to transmit short-range wireless broadcast signals while a short-range wireless connection between the wearable device and a node of the network of nodes is active. A network node initiates the establishment of an internal transmission channel or path within the network for routing sound recorded by a microphone of the wearable device as audio data to the network node assigned to receive incoming, to the network, and forward outgoing, from the network, audio data from a conference call or call in which a particular wearable device is to participate. The network node mutes the microphone or audio data obtained by the microphone in response to detecting certain voice activity in audio data incoming to the network node.

公开(公告)号：US12080308B2

公开(公告)日：2024-09-03

申请号：US17372295

申请日：2021-07-09

申请人： DOLBY LABORATORIES LICENSING CORPORATION ,  DOLBY INTERNATIONAL AB

发明人： Jeroen Koppens ,  Scott Gregory Norcross

IPC分类号： G10L19/00 ,  G10L19/16 ,  G10L19/24 ,  G10L21/034

CPC分类号： G10L19/167 ,  G10L19/24 ,  G10L21/034

摘要： This disclosure falls into the field of audio coding, in particular it is related to the field of providing a framework for providing loudness consistency among differing audio output signals. In particular, the disclosure relates to methods, computer program products and apparatus for encoding and decoding of audio data bitstreams in order to attain a desired loudness level of an output audio signal.

公开(公告)号：US20240221737A1

公开(公告)日：2024-07-04

申请号：US18609542

申请日：2024-03-19

申请人： Google LLC

发明人： Diego Melendo Casado ,  Ignacio Lopez Moreno ,  Javier Gonzalez-Dominguez

IPC分类号： G10L15/20 ,  G06F3/16 ,  G10L15/22 ,  G10L15/26 ,  G10L17/00 ,  G10L17/06 ,  G10L21/034 ,  G10L25/84 ,  H03G3/30

CPC分类号： G10L15/20 ,  G06F3/165 ,  G06F3/167 ,  G10L15/222 ,  G10L17/06 ,  G10L21/034 ,  G10L25/84 ,  H03G3/3005 ,  G10L15/26 ,  G10L17/00

摘要： The technology described in this document can be embodied in a computer-implemented method that includes receiving, at a processing system, a first signal including an output of a speaker device and an additional audio signal. The method also includes determining, by the processing system, based at least in part on a model trained to identify the output of the speaker device, that the additional audio signal corresponds to an utterance of a user. The method further includes initiating a reduction in an audio output level of the speaker device based on determining that the additional audio signal corresponds to the utterance of the user.

公开(公告)号：US12014748B1

公开(公告)日：2024-06-18

申请号：US16988423

申请日：2020-08-07

申请人： Amazon Technologies, Inc.

发明人： Ritwik Giri ,  Mehmet Umut Isik ,  Neerad Dilip Phansalkar ,  Jean-Marc Valin ,  Karim Helwani ,  Arvindh Krishnaswamy

IPC分类号： G10L21/0208 ,  G06N5/04 ,  G06N20/00 ,  G10L21/034

CPC分类号： G10L21/0208 ,  G06N5/04 ,  G06N20/00 ,  G10L21/034 ,  G10L2021/02082

摘要： Techniques for training and using a machine learning model for estimation of reverberation in a multi-task learning framework are described. According to some embodiments, the multi-task learning framework improves the performance of the machine learning model by estimating the amount of reverberation present in an input audio recording as a secondary task to the primary task of generating a clean speech portion of the input audio recording. In one embodiment, a model architecture is selected that takes a noisy reverberant recording as an input and outputs an estimate of a clean (e.g., de-reverberated) signal, an estimate of noise (e.g., background noise), and an estimate of the reverb only portion, with the secondary task of estimating the reverb only portion acting as a regularizer that improves the machine learning model's performance in enhancing the reverberant (e.g., and noisy) input speech.

公开(公告)号：US20240194216A1

公开(公告)日：2024-06-13

申请号：US18582177

申请日：2024-02-20

申请人： Bose Corporation

发明人： Ankita D. Jain ,  Cristian Hera

IPC分类号： G10L21/0364 ,  G10L21/034 ,  G10L25/78 ,  H03G3/30 ,  H04R3/00

CPC分类号： G10L21/0364 ,  G10L21/034 ,  G10L25/78 ,  H04R3/00 ,  H03G3/3005 ,  H04R2430/01 ,  H04R2499/13

摘要： A method for adjusting the clarity of an audio output in a changing environment, including: receiving a content signal; applying a customized gain to the content signal; and outputting the content signal with the customized gain to at least one speaker for transduction to an acoustic signal, wherein the customized gain is applied on a per frequency bin basis such that frequencies of a lesser magnitude are enhanced with respect to frequencies of a greater magnitude and an intelligibility of the acoustic signal is set approximately at a desired level, wherein the customized gain is determined according to at least one of a gain applied to the content signal, a bandwidth of the content signal, and a content type encoded by the content signal.

公开(公告)号：US11930322B2

公开(公告)日：2024-03-12

申请号：US17230680

申请日：2021-04-14

申请人： Orcam Technologies Ltd.

发明人： Roi Nathan ,  Yonatan Wexler ,  Amnon Shashua ,  Tal Rosenwein ,  Oren Tadmor

IPC分类号： H04R25/00 ,  G03B31/00 ,  G06F1/16 ,  G06F3/16 ,  G06F18/21 ,  G06F18/25 ,  G06V10/80 ,  G06V20/10 ,  G06V40/10 ,  G06V40/16 ,  G06V40/20 ,  G10L15/26 ,  G10L17/00 ,  G10L17/04 ,  G10L17/06 ,  G10L17/18 ,  G10L21/003 ,  G10L21/0272 ,  G10L21/034 ,  G10L25/51 ,  H04N5/38 ,  H04N7/18 ,  H04N23/51 ,  H04R1/08 ,  G06F18/00

CPC分类号： H04R25/407 ,  G03B31/00 ,  G06F1/163 ,  G06F1/1686 ,  G06F3/165 ,  G06F3/167 ,  G06F18/21 ,  G06F18/251 ,  G06V10/803 ,  G06V20/10 ,  G06V40/10 ,  G06V40/16 ,  G06V40/165 ,  G06V40/171 ,  G06V40/172 ,  G06V40/20 ,  G10L15/26 ,  G10L17/00 ,  G10L17/04 ,  G10L17/06 ,  G10L17/18 ,  G10L21/003 ,  G10L21/0272 ,  G10L21/034 ,  G10L25/51 ,  H04N5/38 ,  H04N7/185 ,  H04N23/51 ,  H04R1/08 ,  H04R25/405 ,  H04R25/45 ,  H04R25/505 ,  H04R25/554 ,  H04R25/558 ,  H04R25/60 ,  H04R25/606 ,  H04R25/65 ,  G06F18/00 ,  H04R2225/025 ,  H04R2225/41 ,  H04R2225/43 ,  H04R2225/55 ,  H04R2460/01 ,  H04R2460/13

摘要： A system may include a wearable camera configured to capture images and a microphone configured to capture sounds. The system may also include a processor programmed to receive audio signals from the microphone and detect, based on analysis of the audio signals, a first audio signal associated with a first time period. The first audio signal may be representative of a voice of a single individual. The processor may also be programmed to detect, based on analysis of the audio signals, a second audio signal associated with a second time period. The second time period may be different from the first time period, and the second audio signal may be representative of overlapping voices of two or more individuals. The processor may further be programmed to selectively condition the first and second audio signals, and cause transmission of the conditioned first audio signal to a hearing interface device.

公开(公告)号：US20240079023A1

公开(公告)日：2024-03-07

申请号：US17929994

申请日：2022-09-06

申请人： Dell Products, L.P.

发明人： Daniel L. Hamlin ,  Srikanth Kondapi ,  Todd Erick Swierk

IPC分类号： G10L21/034 ,  G06F9/54 ,  G10L21/0272 ,  G10L25/30

CPC分类号： G10L21/034 ,  G06F9/546 ,  G10L21/0272 ,  G10L25/30

摘要： Systems and methods for equalizing audio during a collaboration session in a heterogenous computing platform are described. In an illustrative, non-limiting embodiment, an Information Handling System (IHS) may include: a heterogeneous computing platform comprising a plurality of devices, and a memory coupled to the heterogeneous computing platform, where the memory comprises a plurality of sets of firmware instructions, where each set of firmware instructions, upon execution by a respective device, enables the respective device to provide a corresponding firmware service, and where at least one of the plurality of devices operates as an orchestrator configured to: receive a policy from an Information Technology Decision Maker (ITDM) or Original Equipment Manufacturer (OEM), and select an audio equalization setting usable during a collaboration session based, at least in part, upon the policy.

公开(公告)号：US20240013799A1

公开(公告)日：2024-01-11

申请号：US18044777

申请日：2021-09-21

申请人： Dolby Laboratories Licensing Corporation ,  DOLBY INTERNATIONAL AB

发明人： Davide Scaini ,  Chunghsin Yeh ,  Giulio Cengarle ,  Mark David de Burgh

IPC分类号： G10L21/0232 ,  G10L21/028 ,  G10L25/18 ,  G10L25/84 ,  G10L21/034 ,  G10L21/0364 ,  G10L25/21

CPC分类号： G10L21/0232 ,  G10L21/028 ,  G10L25/18 ,  G10L25/84 ,  G10L21/034 ,  G10L21/0364 ,  G10L25/21

摘要： In some embodiments, a method, comprises: dividing, using at least one processor, an audio input into speech and non-speech segments; for each frame in each non-speech segment, estimating, using the at least one processor, a time-varying noise spectrum of the non-speech segment; for each frame in each speech segment, estimating, using the at least one processor, speech spectrum of the speech segment; for each frame in each speech segment, identifying one or more non-speech frequency components in the speech spectrum; comparing the one or more non-speech frequency components with one or more corresponding frequency components in a plurality of estimated noise spectra and selecting the estimated noise spectrum from the plurality of estimated noise spectra based on a result of the comparing.

公开(公告)号：US20230267947A1

公开(公告)日：2023-08-24

申请号：US18007005

申请日：2021-08-02

申请人： Dolby Laboratories Licensing Corporation

发明人： Zhiwei SHUANG

IPC分类号： G10L21/0364 ,  G10L25/84 ,  G10L21/0232 ,  G10L25/18 ,  G10L21/034

CPC分类号： G10L21/0364 ,  G10L25/84 ,  G10L21/0232 ,  G10L25/18 ,  G10L21/034 ,  G10L25/30

摘要： A method of noise reduction includes using a neural network to control a Wiener filter. The gains estimated by the neural network are combined with the gains produced by the Wiener filter. In this manner, the noise reduction system provides improved results as compared to using only a neural network.

公开(公告)号：US11715483B2

公开(公告)日：2023-08-01

申请号：US16899474

申请日：2020-06-11

申请人： Apple Inc.

发明人： Yang Lu ,  Fatos Myftari ,  Vasu Iyengar ,  Aram M. Lindahl

IPC分类号： G10L21/034 ,  G10L21/0208 ,  G10K11/178 ,  H03G3/30 ,  H04R1/10

CPC分类号： G10L21/034 ,  G10K11/17853 ,  G10L21/0208 ,  H03G3/301 ,  H04R1/1091 ,  G10K2210/1081 ,  G10K2210/3218 ,  H04R2460/01

摘要： Aspects of the subject technology relate to a device including a microphone, a filter and a processor. The filter receives an audio signal including ambient noise and a voice of a user of the device from the microphone. At least a portion of ambient noise is filtered from the audio signal. The processor determines a level of the ambient noise in the received audio signal and dynamically adjusts a gain applied to the filtered audio signal based on the level of the ambient noise.