Neural Networks For Speaker Verification

Patent metadata

Field	Value
Publication number	US-2024038245-A1
Application number	US-202318485069-A
Country	US
Kind code	A1
Filing date	Oct 11, 2023
Priority date	Sep 4, 2015
Publication date	Feb 1, 2024
Grant date	—

Abstract

This document generally describes systems, methods, devices, and other techniques related to speaker verification, including (i) training a neural network for a speaker verification model, (ii) enrolling users at a client device, and (iii) verifying identities of users based on characteristics of the users' voices. Some implementations include a computer-implemented method. The method can include receiving, at a computing device, data that characterizes an utterance of a user of the computing device. A speaker representation can be generated, at the computing device, for the utterance using a neural network on the computing device. The neural network can be trained based on a plurality of training samples that each: (i) include data that characterizes a first utterance and data that characterizes one or more second utterances, and (ii) are labeled as a matching speakers sample or a non-matching speakers sample.

First claim

What is claimed is:

1 . A computer-implemented method executed on data processing hardware that causes the data processing hardware to perform operations comprising:
receiving a plurality of training samples, each training sample comprising:
a training verification utterance;
a training enrollment utterance spoken by a corresponding speaker; and
a first classification for the training sample that indicates whether a speaker of the training utterance is the same or different from the corresponding speaker that spoke the training enrollment utterance;
training a neural network on the plurality of training samples by:
for each training sample:
processing, using the neural network, audio signals characterizing the training verification utterance to generate a first training speaker representation for the training verification utterance;
processing, using the neural network, the audio signals characterizing the training enrollment utterance;
determining a second classification for the training sample based on the first training speaker representation and the second training speaker representation, the second classification for the training sample indicating whether the speaker of the training utterance is the same or different from the corresponding speaker that spoke the training enrollment utterance; and
adjusting parameters of the neural network based on a comparison of the first classification of the training sample and the second classification determined for the training sample; and

transmitting the trained neural network over a network to a user device.

2 . The computer-implemented method of claim 1 , wherein the user device is configured to:
receive audio signals that represent the user speaking the plurality of enrollment utterances, the audio signals representing the user speaking the plurality of enrollment utterances recorded by the user device; generate, using the trained neural configured to receive the audio signals representing the user speaking the plurality of enrollment utterances as input, a reference speaker model associated with the user that characterizes distinctive features of a voice of the user; and store the reference speaker model on memory hardware of the user device.

3 . The computer-implemented method of claim 2 , wherein the trained neural network comprises a trained neural network comprising a long short-term memory (LSTM) layer configured to receive the audio signals representing the user speaking the plurality of enrollment utterances as input.

4 . The computer-implemented method of claim 3 , wherein the user device is further configured to:
obtain a plurality of audio frames representing a first utterance; generate, using the trained neural network, a speaker representation for the utterance, the speaker representation indicating distinctive features of a speaker of the first utterance; determine a similarity score between the speaker representation for the first utterance and the reference speaker model stored on the memory hardware of the user device satisfies a similarity score threshold; and authenticate the speaker of the first utterance as the user associated with the reference speaker model based on determining the similarity score satisfies the similarity score threshold.

5 . The computer-implemented method of claim 4 , wherein the trained neural network further comprises a fully-connected linear layer configured to:
receive, as input, an output of the LSTM layer; and generate, as output, the speaker representation for the first utterance.

6 . The computer-implemented method of claim 4 , wherein the user device is further configured to, in response to authenticating the speaker of the first utterance as the user associated with the reference speaker model, transition operation of the user device from a low-power state to a more fully-featured state.

7 . The computer-implemented method of claim 1 , wherein the training verification utterance and the training enrollment utterance comprise a same pre-determined phrase.

8 . The computer-implemented method of claim 1 , wherein the comparison of the first classification of the training sample and the second classification determined for the training sample comprises a cosine distance between a vector of values for the first classification and a vector of values for the second classification.

9 . The computer-implemented method of claim 1 , wherein the user device comprises a smart phone.

10 . The computer-implemented method of claim 1 , wherein the data processing hardware resides on a remote computing device.

11 . A system comprising:
data processing hardware; and memory hardware in communication with the data processing hardware and storing instructions that when executed by the data processing hardware cause the data processing hardware to perform operations comprising:
receiving a plurality of training samples, each training sample comprising:
a training verification utterance;
a training enrollment utterance spoken by a corresponding speaker; and
a first classification for the training sample that indicates whether a speaker of the training utterance is the same or different from the corresponding speaker that spoke the training enrollment utterance;

training a neural network on the plurality of training samples by:
for each training sample:
processing, using the neural network, audio signals characterizing the training verification utterance to generate a first training speaker representation for the training verification utterance;
processing, using the neural network, the audio signals characterizing the training enrollment utterance;
determining a second classification for the training sample based on the first training speaker representation and the second training speaker representation, the second classification for the training sample indicating whether the speaker of the training utterance is the same or different from the corresponding speaker that spoke the training enrollment utterance; and
adjusting parameters of the neural network based on a comparison of the first classification of the training sample and the second classification determined for the training sample; and

transmitting the trained neural network over a network to a user device.

12 . The system of claim 11 , wherein the user device is configured to:
receive audio signals that represent the user speaking the plurality of enrollment utterances, the audio signals representing the user speaking the plurality of enrollment utterances recorded by the user device; generate, using the trained neural configured to receive the audio signals representing the user speaking the plurality of enrollment utterances as input, a reference speaker model associated with the user that characterizes distinctive features of a voice of the user; and store the reference speaker model on memory hardware of the user device.

13 . The system of claim 12 , wherein the trained neural network comprises a trained neural network comprising a long short-term memory (LSTM) layer configured to receive the audio signals representing the user speaking the plurality of enrollment utterances as input.

14 . The system of claim 13 , wherein the user device is further configured to:
obtain a plurality of audio frames representing a first utterance; generate, using the trained neural network, a speaker representation for the utterance, the speaker representation indicating distinctive features of a speaker of the first utterance; determine a similarity score between the speaker representation for the first utterance and the refere

Assignees

• Google Llc

Inventors

• Heigold Georg
• Bengio Samuel
• Lopez Moreno Ignacio

Classifications

• G10L17/18 Primary

  Artificial neural networks; Connectionist approaches · CPC title

• G10L17/04

  Training, enrolment or model building · CPC title

• G10L17/02 Primary

  Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction · CPC title

• G07C9/37 Primary

  using biometric data, e.g. fingerprints, iris scans or voice recognition · CPC title