US 11,915,687 B1Granted

Systems and methods for generating labeled data to facilitate configuration of network microphone devices

AssigneeSonos, Inc.

InventorsConnor Kristopher Smith, Kurt Thomas Soto, Charles Conor Sleith

Life of this patent

Estimating the term…

Loading patent…

Abstract

Systems and methods for generating training data are described herein. Pieces of metadata captured by a plurality of networked sensor systems can be captured, where each piece of metadata is associated with a specific set of sensor data captured by one of the plurality of networked sensor systems and includes a set of characteristics for the specific set of captured sensor data. A probabilistic model can be generated based on the received metadata and simulations can be performed based upon a training corpus by generating multiple scenarios, and, for each scenario, a scenario specific version of a particular annotated sample is generated by performing a simulation using the particular annotated sample. The scenario specific versions of annotated samples from the training corpus can be stored as a training data set on the at least one network device.

Claims24

Dependent claims

A computer system comprising:

a communication interface configured to facilitate communication via at least one data network;
at least one processor;
and
at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor such that the computer system is configured to:
receive a plurality of sets of sound data captured using a plurality of network microphone devices (NMDs), where each of the NMDs is configured in accordance with a set of NMD software configuration parameters;
receive, for each captured set of sound data, sound metadata associated with a respective captured set of sound data, where each piece of sound metadata comprises a set of characteristics for a set of sound data;
generate a probabilistic model based on the received sound metadata;
perform a plurality of acoustic simulations to obtain a noised version of a training data set of annotated speech samples at least in part by:
generating a plurality of scenarios, wherein, for a scenario of the plurality of scenarios:
the scenario comprises a set of characteristics for a set of sound data;
and
values for characteristics in the set of characteristics for the scenario are drawn from the probabilistic model;
and
for each of the plurality of scenarios, generating a noised version of a specific annotated speech sample by performing an acoustic simulation based upon a specific scenario from the plurality of scenarios;
simulate performance of a set of modified NMD software configuration parameters using the noised version of the training data set of annotated speech samples;
and
cause software on at least one NMD to be updated based on the set of modified NMD software configuration parameters.

14 more elements

A method for updating software configuration parameters of at least one network microphone device (NMD), the method comprising:

receiving a plurality of sets of sound data captured using a plurality of network microphone devices (NMDs), where each of the NMDs is configured in accordance with a set of NMD software configuration parameters;
receiving, for each captured set of sound data, sound metadata associated with a respective captured set of sound data, where each piece of sound metadata comprises a set of characteristics for a set of sound data;
generating a probabilistic model based on the received sound metadata;
performing a plurality of acoustic simulations to obtain a noised version of a training data set of annotated speech samples, wherein performing the plurality of acoustic simulations comprises:
generating a plurality of scenarios, wherein, for a scenario of the plurality of scenarios:
the scenario comprises a set of characteristics for a set of sound data;
and
values for characteristics in the set of characteristics for the scenario are drawn from the probabilistic model;
and
for each of the plurality of scenarios, generating a noised version of a specific annotated speech sample by performing an acoustic simulation based upon a specific scenario from the plurality of scenarios;
simulating performance of a set of modified NMD software configuration parameters using the noised version of the training data set of annotated speech samples;
and
causing software on at least one NMD to be updated based on the set of modified NMD software configuration parameters.

10 more elements

A non-transitory computer-readable medium, wherein the non-transitory computer-readable medium is provisioned with program instructions that, when executed by at least one processor, cause a computing system to:

receive a plurality of sets of sound data captured using a plurality of network microphone devices (NMDs), where each of the NMDs is configured in accordance with a set of NMD software configuration parameters;
receive, for each captured set of sound data, sound metadata associated with a respective captured set of sound data, where each piece of sound metadata comprises a set of characteristics for a set of sound data;
generate a probabilistic model based on the received sound metadata;
perform a plurality of acoustic simulations to obtain a noised version of a training data set of annotated speech samples at least in part by:
generating a plurality of scenarios, wherein, for a scenario of the plurality of scenarios:
the scenario comprises a set of characteristics for a set of sound data;
and
values for characteristics in the set of characteristics for the scenario are drawn from the probabilistic model;
and
for each of the plurality of scenarios, generating a noised version of a specific annotated speech sample by performing an acoustic simulation based upon a specific scenario from the plurality of scenarios;
simulate performance of a set of modified NMD software configuration parameters using the noised version of the training data set of annotated speech samples;
and
cause software on at least one NMD to be updated based on the set of modified NMD software configuration parameters.

10 more elements

Dependencies are inferred from the claim text.

Facts

Application: 18/152,096Filed 9 Jan 2023 · continuation
Priority: 26 Sep 2019
Granted: 27 Feb 2024
Term ends: Not estimatedThe source records don’t support an estimate.

Classification13 codes

G10L 15/063
Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice: training
G10L 15/148
Hidden Markov Models [HMMs]: duration modelling in HMMs, e.g. semi HMM, segmental models or transition probabilities
G10L 15/16
Speech classification or search: using artificial neural networks
G10L 15/22
Speech recognition: procedures used during a speech recognition process, e.g. man-machine dialogue