A client device processing received emoji messages using emoji-first messaging. Text messaging is automatically converted to emojis by an emoji-first application so that only emojis are communicated from one client device to another client device. Each client device has a library of emojis that are mapped to words, which libraries are customizable and unique to the users of the client devices, such that the users can communicate secretly in code. Upon receipt of a string of emojis, a user can select the emoji string to convert to text if desired, for a predetermined period of time.

A method includes receiving a sequence of acoustic frames and generating, by a first encoder, a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The method also includes generating, by a first pass transducer decoder, a first pass speech recognition hypothesis for a corresponding first higher order feature representation and generating, by a text encoder, a text encoding for a corresponding first pass speech recognition hypothesis. The method also includes generating, by a second encoder, a second higher order feature representation for a corresponding first higher order feature representation. The method also includes generating, by a second pass transducer decoder, a second pass speech recognition hypothesis using a corresponding second higher order feature representation and a corresponding text encoding.

A method includes receiving a sequence of acoustic frames and generating, by a first encoder, a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The method also includes generating, by a first pass transducer decoder, a first pass speech recognition hypothesis for a corresponding first higher order feature representation and generating, by a text encoder, a text encoding for a corresponding first pass speech recognition hypothesis. The method also includes generating, by a second encoder, a second higher order feature representation for a corresponding first higher order feature representation. The method also includes generating, by a second pass transducer decoder, a second pass speech recognition hypothesis using a corresponding second higher order feature representation and a corresponding text encoding.

A method for event extraction according to the disclosure includes: processing an object text using a preset extraction model to determine event information of the object text; wherein the event information includes an event element, and an event type and a role corresponding to the event element; and the extraction model includes a classification layer and an output layer; the classification layer is configured to determine a token attribute of a token in the object text; the token attribute includes whether the token is a start token of the event element of any event type and any role, and whether the token is an end token of the event element of any event type and any role; and the output layer is configured to determine the event element according to the token attribute of the token, and determine the event type and the role corresponding to the event element.

A method for event extraction according to the disclosure includes: processing an object text using a preset extraction model to determine event information of the object text; wherein the event information includes an event element, and an event type and a role corresponding to the event element; and the extraction model includes a classification layer and an output layer; the classification layer is configured to determine a token attribute of a token in the object text; the token attribute includes whether the token is a start token of the event element of any event type and any role, and whether the token is an end token of the event element of any event type and any role; and the output layer is configured to determine the event element according to the token attribute of the token, and determine the event type and the role corresponding to the event element.

Unicode data can be protected in a distributed tokenization environment. Data to be tokenized can be accessed or received by a security server, which instantiates a number of tokenization pipelines for parallel tokenization of the data. Unicode token tables are accessed by the security server, and each tokenization pipeline uses the accessed token tables to tokenization a portion of the data. Each tokenization pipeline performs a set of encoding or tokenization operations in parallel and based at least in part on a value received from another tokenization pipeline. The outputs of the tokenization pipelines are combined, producing tokenized data, which can be provided to a remote computing system for storage or processing.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using memory-optimized contrastive learning to train image encoder and text encoder neural networks.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using memory-optimized contrastive learning to train image encoder and text encoder neural networks.

A computer-implemented method of encoding a word for use in a method of text analysis comprises receiving input text to be analysed, the input text comprising a first word which is not represented in a vocabulary set stored on a storage. The vocabulary set comprises a plurality of words and an associated word embedding vector for each word in the set. The method comprises identifying the first word as a word which is not represented in the vocabulary set and determining one or more sub-words within the first word with which to encode the first word. Each of the one or more sub-words corresponds with a word represented in the vocabulary set and having an embedding vector in the vocabulary set. The method comprises determining an encoding for the first word based on the one or more sub-words.

A computer-implemented method of encoding a word for use in a method of text analysis comprises receiving input text to be analysed, the input text comprising a first word which is not represented in a vocabulary set stored on a storage. The vocabulary set comprises a plurality of words and an associated word embedding vector for each word in the set. The method comprises identifying the first word as a word which is not represented in the vocabulary set and determining one or more sub-words within the first word with which to encode the first word. Each of the one or more sub-words corresponds with a word represented in the vocabulary set and having an embedding vector in the vocabulary set. The method comprises determining an encoding for the first word based on the one or more sub-words.