Abstract: A method for performing speech recognition using sequence-to-sequence models includes receiving audio data for an utterance and providing features indicative of acoustic characteristics of the utterance as input to an encoder. The method also includes processing an output of the encoder using an attender to generate a context vector, generating speech recognition scores using the context vector and a decoder trained using a training process, and generating a transcription for the utterance using word elements selected based on the speech recognition scores. The transcription is provided as an output of the ASR system.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer-readable media, for speech recognition using multi-dialect and multilingual models. In some implementations, audio data indicating audio characteristics of an utterance is received. Input features determined based on the audio data are provided to a speech recognition model that has been trained to output score indicating the likelihood of linguistic units for each of multiple different language or dialects. The speech recognition model can be one that has been trained using cluster adaptive training. Output that the speech recognition model generated in response to receiving the input features determined based on the audio data is received. A transcription of the utterance generated based on the output of the speech recognition model is provided.

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: Methods, systems, and apparatus, including computer programs encoded on computer-readable storage media, for speech recognition using attention-based sequence-to-sequence models. In some implementations, audio data indicating acoustic characteristics of an utterance is received. A sequence of feature vectors indicative of the acoustic characteristics of the utterance is generated. The sequence of feature vectors is processed using a speech recognition model that has been trained using a loss function that uses a set of speech recognition hypothesis samples, the speech recognition model including an encoder, an attention module, and a decoder. The encoder and decoder each include one or more recurrent neural network layers. A sequence of output vectors representing distributions over a predetermined set of linguistic units is obtained. A transcription for the utterance is obtained based on the sequence of output vectors. Data indicating the transcription of the utterance is provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer-readable media, for speech recognition using multi-dialect and multilingual models. In some implementations, audio data indicating audio characteristics of an utterance is received. Input features determined based on the audio data are provided to a speech recognition model that has been trained to output score indicating the likelihood of linguistic units for each of multiple different language or dialects. The speech recognition model can be one that has been trained using cluster adaptive training. Output that the speech recognition model generated in response to receiving the input features determined based on the audio data is received. A transcription of the utterance generated based on the output of the speech recognition model is provided.

Abstract: Optical center is determined on a column-by-column and row-by-row basis by identifying brightest pixels in respective columns and rows. The brightest pixels in each column are identified and a line is fit to those pixels. Similarly, brightest pixels in each row are identified and a second line is fit to those pixels. The intersection of the two lines is the optical center.

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: The invention provides a laboratory apparatus for the storage and delivery of samples or labware wherein the apparatus can have a plurality of beams for holding one or more containers of samples or labware consumables in a first set position; a conveyor for motional translation of the one or more supports from a first set position to a second set position where the conveyor has a first pulley in a first set of pulleys translationally connected by a first timing belt to a second pulley in the first set of pulleys, and a first pulley in a second set of pulleys translationally connected by a second timing belt to a second pulley in the second set of pulleys.

Abstract: Optical center is determined on a column-by-column and row-by-row basis by identifying brightest pixels in respective columns and rows. The brightest pixels in each column are identified and a line is fit to those pixels. Similarly, brightest pixels in each row are identified and a second line is fit to those pixels. The intersection of the two lines is the optical center.

Abstract: A computer-implemented system including a data store: a content items database, a user account database, and one or more servers configured to execute a computer program to perform one or more of the following: receiving content items from content database that are associated with a topic selected by a user for posting on a social network, wherein at least one content item is associated with an URL; estimating a post to a reaction filter for a time interval for the social network for the user, calculating a reaction profile associated with reactions to posts on the social network by aggregating reaction time of a plurality of users on the social network for one or more content items posted on the social network; determining a schedule for posting the content items on the social network as a function of the post to reaction filter and reaction profile.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-readable storage media, for speech recognition using attention-based sequence-to-sequence models. In some implementations, audio data indicating acoustic characteristics of an utterance is received. A sequence of feature vectors indicative of the acoustic characteristics of the utterance is generated. The sequence of feature vectors is processed using a speech recognition model that has been trained using a loss function that uses a set of speech recognition hypothesis samples, the speech recognition model including an encoder, an attention module, and a decoder. The encoder and decoder each include one or more recurrent neural network layers. A sequence of output vectors representing distributions over a predetermined set of linguistic units is obtained. A transcription for the utterance is obtained based on the sequence of output vectors. Data indicating the transcription of the utterance is provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-readable storage media, for speech recognition using attention-based sequence-to-sequence models. In some implementations, audio data indicating acoustic characteristics of an utterance is received. A sequence of feature vectors indicative of the acoustic characteristics of the utterance is generated. The sequence of feature vectors is processed using a speech recognition model that has been trained using a loss function that uses N-best lists of decoded hypotheses, the speech recognition model including an encoder, an attention module, and a decoder. The encoder and decoder each include one or more recurrent neural network layers. A sequence of output vectors representing distributions over a predetermined set of linguistic units is obtained. A transcription for the utterance is obtained based on the sequence of output vectors. Data indicating the transcription of the utterance is provided.

Abstract: A computer-implemented system including a data store: a content items database, a user account database, and one or more servers configured to execute a computer program to perform one or more of the following: receiving content items from content database that are associated with a topic selected by a user for posting on a social network, wherein at least one content item is associated with an URL; estimating a post to a reaction filter for a time interval for the social network for the user, calculating a reaction profile associated with reactions to posts on the social network by aggregating reaction time of a plurality of users on the social network for one or more content items posted on the social network; determining a schedule for posting the content items on the social network as a function of the post to reaction filter and reaction profile.

Abstract: A virtual make-up apparatus and method: store cosmetic item information of cosmetic items of different colors; store a different texture component for each stored cosmetic item of a specific color; extract an object portion image of a virtual make-up from a facial image; extract color information from the object portion image; designate an item of the virtual make-up corresponding to a stored cosmetic item and output a color image by applying a color corresponding to the designated item on the object portion image; output a texture image, based on analyzed color information corresponding to a stored cosmetic item, by adding a texture component to a part of the object portion image; and display a virtual make-up image of virtual make-up using the designated item applied on the facial image, by using the color and texture images, and the object portion image of the virtual make-up of the facial image.

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 terminal images first and second images respectively indicating facial images of a user before and after makeup, acquires information on a type or region of the makeup performed by the user, and transmits the first and second images and the information on the type or region of the makeup in association with each other to a server. The server deduces a makeup color of the makeup performed by the user based on the first and second images and the information on the type or region of the makeup performed by the user, and extracts at least one similar makeup item having the makeup color based on information on the makeup color and a makeup item database, and transmits information on at least one similar makeup item to the terminal. A terminal displays information on at least one similar makeup item transmitted from the server to a display unit.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing point cloud data representing a sensor measurement of a scene captured by one or more sensors to generate an object detection output that identifies locations of one or more objects in the scene. When deployed within an on-board system of a vehicle, the object detection output that is generated can be used to make autonomous driving decisions for the vehicle with enhanced accuracy.

Abstract: Alternating Current (AC) light sources can cause images captured using a rolling shutter to include alternating darker and brighter regions—known as flicker bands—due to some sensor rows being exposed to different intensities of light than others. Flicker bands may be compensated for by extracting them from images that are captured using exposures that at least partially overlap in time. Due to the overlap, the images may be subtracted from each other so that scene content substantially cancels out, leaving behind flicker bands. The images may be for a same frame captured by at least one sensor, such as different exposures for a frame. For example, the images used to extract flicker bands may be captured using different exposure times that share a common start time, such as using a multi-exposure sensor where light values are read out at different times during light integration.

Abstract: Alternating Current (AC) light sources can cause images captured using a rolling shutter to include alternating darker and brighter regions—known as flicker bands—due to some sensor rows being exposed to different intensities of light than others. Flicker bands may be compensated for by extracting them from images that are captured using exposures that at least partially overlap in time. Due to the overlap, the images may be subtracted from each other so that scene content substantially cancels out, leaving behind flicker bands. The images may be for a same frame captured by at least one sensor, such as different exposures for a frame. For example, the images used to extract flicker bands may be captured using different exposure times that share a common start time, such as using a multi-exposure sensor where light values are read out at different times during light integration.

Abstract: Optical center is determined on a column-by-column and row-by-row basis by identifying brightest pixels in respective columns and rows. The brightest pixels in each column are identified and a line is fit to those pixels. Similarly, brightest pixels in each row are identified and a second line is fit to those pixels. The intersection of the two lines is the optical center.