Abstract: A method of generating an accurate speaker representation for an audio sample includes receiving a first audio sample from a first speaker and a second audio sample from a second speaker. The method includes dividing a respective audio sample into a plurality of audio slices. The method also includes, based on the plurality of slices, generating a set of candidate acoustic embeddings where each candidate acoustic embedding includes a vector representation of acoustic features. The method further includes removing a subset of the candidate acoustic embeddings from the set of candidate acoustic embeddings. The method additionally includes generating an aggregate acoustic embedding from the remaining candidate acoustic embeddings in the set of candidate acoustic embeddings after removing the subset of the candidate acoustic embeddings.

Abstract: Embodiments of the invention are directed to a method. The method may include transmitting, by a first device, an encrypted first biometric template generated from a first biometric sample of a user of the first device to a second device, wherein the second device inputs the encrypted first biometric template and a second biometric template generated from a second biometric sample of the user into a function to generate an encoded output. The first device may receive the encoded output from the second device, and may decode the encoded output to recover the encrypted first biometric template and the second biometric template of the user. Upon determining a match result between first and second biometric templates, the first device may transmit unique data to the second device.

Abstract: A method is disclosed. The method includes receiving a broadcast signal from a beacon device, the broadcast signal encoding a first credential associated with a first entity. In response to receipt of the broadcast signal, the mobile communication device transmits the received first credential to an authentication system. The authentication system determines if the first entity associated with the broadcast signal is authentic and generates a confirmation message confirming the authenticity of the first entity. The mobile communication device then receives the confirmation message indicating that the first entity is authentic. The mobile communication thereafter receives and transmits a second credential for the mobile communication device to the beacon device, which transmits the second credential to the authentication system. The authentication system then confirms the authenticity of the mobile communication device.

Abstract: Techniques are described for training and/or utilizing an end-to-end speaker diarization model. In various implementations, the model is a recurrent neural network (RNN) model, such as an RNN model that includes at least one memory layer, such as a long short-term memory (LSTM) layer. Audio features of audio data can be applied as input to an end-to-end speaker diarization model trained according to implementations disclosed herein, and the model utilized to process the audio features to generate, as direct output over the model, speaker diarization results. Further, the end-to-end speaker diarization model can be a sequence-to-sequence model, where the sequence can have variable length. Accordingly, the model can be utilized to generate speaker diarization results for any of various length audio segments.

Abstract: A carbon nanotube field emitter comprises at least two electrodes and at least one graphitized carbon nanotube structure. The at least one graphitized carbon nanotube structure comprises a first end and a field emission end. The first end is opposite to the field emission end. The first end is fixed between the at least two electrodes, and the field emission end is exposed from the at least two electrodes and configured to emit electrons.

Abstract: Embodiments of the invention are directed assessing reliability between two computing devices. A distributed database may maintain reliability associations between pairs of computing devices. Each reliability association may indicate a particular device has determined (e.g., locally) that another device is reliable. In order to determine an amount of reliability between a first computing device and a second computing device, an ordered combination of the reliability associations may be determined utilizing the distributed database. The ordered combination of reliability associations may identify a reliability path between the first computing device and the second computing device. An amount of reliability may be determined based on the reliability path. An interaction between the devices may be allowed or restricted based at least in part on the amount of reliability between the computing devices.

Abstract: A method for using beauty instrument with mask is provided. The method comprises providing a beauty instrument with mask comprising a flexible mask and a controller, applying the flexible mask of on a user's face, and turning on the controller and selecting a function button on the controller, inputting a current to a plurality of functional layers in the flexible mask, and stimulating face skin with the current.

Abstract: Text independent speaker recognition models can be utilized by an automated assistant to verify a particular user spoke a spoken utterance and/or to identify the user who spoke a spoken utterance. Implementations can include automatically updating a speaker embedding for a particular user based on previous utterances by the particular user. Additionally or alternatively, implementations can include verifying a particular user spoke a spoken utterance using output generated by both a text independent speaker recognition model as well as a text dependent speaker recognition model. Furthermore, implementations can additionally or alternatively include prefetching content for several users associated with a spoken utterance prior to determining which user spoke the spoken utterance.

Abstract: The present disclosure provides a broadband dual-polarized solar cell antenna and an antenna array. The broadband dual-polarized solar cell antenna includes an antenna dipole layer, an isolation layer, a solar cell layer, and a ground that are arranged sequentially from top to bottom, where the antenna dipole layer is connected to the ground and a radio frequency (RF) coaxial connector through a metal feeding probe structure, the solar cell layer is placed on the ground, the isolation layer is located between the antenna dipole layer and the solar cell layer, and the isolation layer is made of a transparent material. The present disclosure is small in sunlight shielding and high in transparency, and has a broadband dual-polarized wide-angle scanning capability, which ensures performance of the antenna and power generation efficiency of the solar cell, and is highly applicable in engineering.

Abstract: A system enables entities to access a single platform in order to utilize electronic data storage for storing different types of information. One or more computers may operate an electronic data storage processing network that entities can access when updating information in electronic data storage. The electronic data storage processing network may operate a plurality of electronic data storage processing modules, which can include an aggregator module, a formatter module, an operator signer module, and a validator module. Based on the specific use case for which electronic data storage is utilized, recordable data that is to be added to the electronic data storage can be processed by the appropriate aggregating, formatting, signing, and validating functions provided by the electronic data storage processing modules.

Abstract: A method for using a mask-type beauty instrument is provided. The method comprises providing a mask-type beauty instrument comprising a flexible mask and a controller, applying the flexible mask of on a user's face, and turning on the controller and selecting a function button on the controller, inputting a current to a plurality of functional layers in the flexible mask, and stimulating face skin with the current.

Abstract: A speaker verification method includes receiving audio data corresponding to an utterance, processing the audio data to generate a reference attentive d-vector representing voice characteristics of the utterance, the evaluation ad-vector includes ne style classes each including a respective value vector concatenated with a corresponding routing vector. The method also includes generating using a self-attention mechanism, at least one multi-condition attention score that indicates a likelihood that the evaluation ad-vector matches a respective reference ad-vector associated with a respective user. The method also includes identifying the speaker of the utterance as the respective user associated with the respective reference ad-vector based on the multi-condition attention score.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for speech synthesis. The methods, systems, and apparatus include actions of obtaining an audio representation of speech of a target speaker, obtaining input text for which speech is to be synthesized in a voice of the target speaker, generating a speaker vector by providing the audio representation to a speaker encoder engine that is trained to distinguish speakers from one another, generating an audio representation of the input text spoken in the voice of the target speaker by providing the input text and the speaker vector to a spectrogram generation engine that is trained using voices of reference speakers to generate audio representations, and providing the audio representation of the input text spoken in the voice of the target speaker for output.

Abstract: Embodiments of the invention are directed to a method. The method may include transmitting, by a first device, an encrypted first biometric template generated from a first biometric sample of a user of the first device to a second device, wherein the second device inputs the encrypted first biometric template and a second biometric template generated from a second biometric sample of the user into a function to generate an encoded output. The first device may receive the encoded output from the second device, and may decode the encoded output to recover the encrypted first biometric template and the second biometric template of the user. Upon determining a match result between first and second biometric templates, the first device may transmit unique data to the second device.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for speech synthesis. The methods, systems, and apparatus include actions of obtaining an audio representation of speech of a target speaker, obtaining input text for which speech is to be synthesized in a voice of the target speaker, generating a speaker vector by providing the audio representation to a speaker encoder engine that is trained to distinguish speakers from one another, generating an audio representation of the input text spoken in the voice of the target speaker by providing the input text and the speaker vector to a spectrogram generation engine that is trained using voices of reference speakers to generate audio representations, and providing the audio representation of the input text spoken in the voice of the target speaker for output.

Abstract: A method includes receiving an overlapped audio signal that includes audio spoken by a speaker that overlaps a segment of synthesized playback audio. The method also includes encoding a sequence of characters that correspond to the synthesized playback audio into a text embedding representation. For each character in the sequence of characters, the method also includes generating a respective cancelation probability using the text embedding representation. The cancelation probability indicates a likelihood that the corresponding character is associated with the segment of the synthesized playback audio overlapped by the audio spoken by the speaker in the overlapped audio signal.

Abstract: Techniques disclosed herein are directed towards streaming keyphrase detection which can be customized to detect one or more particular keyphrases, without requiring retraining of any model(s) for those particular keyphrase(s). Many implementations include processing audio data using a speaker separation model to generate separated audio data which isolates an utterance spoken by a human speaker from one or more additional sounds not spoken by the human speaker, and processing the separated audio data using a text independent speaker identification model to determine whether a verified and/or registered user spoke a spoken utterance captured in the audio data. Various implementations include processing the audio data and/or the separated audio data using an automatic speech recognition model to generate a text representation of the utterance.

Abstract: Provided are a method and apparatus for generating a three-dimensional model. The method includes following. A first image containing a first face is acquired. First point cloud data including contour information of the first face is determined based on the first image. First albedo information of the first face and second point cloud data including detail information of the first face are determined based on the first point cloud data and the first image. A three-dimensional model of the first face is generated based on the first albedo information and the second point cloud data.

Abstract: Embodiments of the invention are directed assessing reliability between two computing devices. A distributed database may maintain reliability associations between pairs of computing devices. Each reliability association may indicate a particular device has determined (e.g., locally) that another device is reliable. In order to determine an amount of reliability between a first computing device and a second computing device, an ordered combination of the reliability associations may be determined utilizing the distributed database. The ordered combination of reliability associations may identify a reliability path between the first computing device and the second computing device. An amount of reliability may be determined based on the reliability path. An interaction between the devices may be allowed or restricted based at least in part on the amount of reliability between the computing devices.

Abstract: Embodiments of the invention are directed to the utilization of trust tokens to perform secure message transactions between two devices. A trust token transmitted in a message from one device may include first data that is digitally signed by a trust provider computer, and second data that is digitally signed by the device itself. Upon receipt of a message containing a trust token, the recipient may utilize the first data to verify with the trust provider computer that the sender of the message is a trusted party. The trust provider computer may provide the recipient device the public key of the sender. The recipient may utilize the second data and the provided public key to verify that the sender signed the message and that the message is unaltered. These techniques may increase detection of relay, replay, or other man-in-the-middle attacks, decreasing the likelihood that such attacks will be successful.