Abstract: Features are disclosed for performing functions in response to user requests. Natural Language Understanding (“NLU”) processing may be performed to generate command data that represents a subject of an utterance. The command data may be sent to an application that causes presentation of first output content in a first modality at a first time in response to receiving the command data, and generates second output content in a second modality different from the first modality, wherein the second output content is associated with the first output content. The second output content may be presented in the second modality at a second time subsequent to the first time.

Abstract: An apparatus for generating an encoded signal includes: a window sequence controller for generating a window sequence information for windowing an audio or image signal, the window sequence information indicating a first window for generating a first frame of spectral values, a second window function and at least one third window function for generating a second frame of spectral values, wherein the first window function, the second window function and the one or more third window functions overlap within a multi-overlap region; a preprocessor for windowing a second block of samples corresponding to the second window function and the at least one third window functions using an auxiliary window function to acquire a second block of windowed samples, a spectrum converter for applying an aliasing-introducing transform; and a processor for processing the first frame and the second frame to acquire encoded frames of the audio or image signal.

Abstract: Neural network systems are provided that comprise one or more neural networks. The first neural network can comprise a convolutional neural network (CNN) long short-term memory (LSTM) architecture for receiving a primary data set comprising text messages and output a primary data structure comprising a text pattern-based feature. The second neural network can comprise a CNN architecture for receiving a secondary data sets derived from the primary data set and output a plurality of secondary data structures. The third neural network can combine the data structures to produce a combined data structure, and then process it to produce a categorized data structure comprising the text messages assigned to targets. The primary data set can comprise hate speech and the categorized data structure can comprise target categories, for example, hate targets. Methods of operating neural network systems and computer program products for performing such methods are also provided.

Abstract: Methods, systems, and computer program products for training an artificial intelligence (AI) of a voice response system. Aspects include receiving, by the voice response system from a user, a voice command to perform a requested action and interpreting, by an AI model, the voice command. Aspects also include performing an action based on the interpretation of the voice command and receiving non-verbal feedback from the user. Aspects further include updating the AI model based on a determination that the non-verbal feedback indicates that the user is not satisfied with the action performed.

Abstract: A natural-language voice chatbot engages a consumer in a voice dialogue. The chatbot is customized for engaging the specific consumer based on features and characteristics of that specific consumer's speech and a lexicon associated with terms, words, and commands for item ordering. The consumer can perform voice queries for specific items and/or specific establishments for placing a pre-staged order with the chatbot. Once the consumer selects options with a specific establishment, a pre-staged order is provided to the corresponding establishment on behalf of the user. Location data for a consumer-operated device is monitored and when it is determined that the consumer will arrive at the establishment within a time period required by the establishment to prepare the pre-staged order, a message is sent to the establishment to begin preparing the pre-staged order.

Abstract: Embodiments described provide a method for mixing vocal performances from different vocalists. A vocal score temporally synchronized with a corresponding backing track and lyrics is retrieved via a communications interface of a portable computing device. A first vocal performance of a user is captured, via a microphone interface of the portable computing device, and in correspondence with the backing track. An open call indication for soliciting, from a second vocalist, a second vocal performance to be mixed for audible rendering with the first vocal performance is transmitted. A mix to one of the user and the second vocalist is provided by selecting, based on to whom the mix is provided, the mix from alternative mixes each having a different prominent vocal performance.

Abstract: This application related to Artificial Intelligence technical field and discloses a robot and a method for predicting an emotion status by a robot. The method includes: determining a first emotion status of a first user, where the first emotion status is an emotion status of the first user at a first moment; predicting a second emotion status based on the first emotion status and a first emotion prediction model, where the second emotion status is an emotion status of the first user at a second moment, and the second moment is later than the first moment; and outputting a response to the first user based on the second emotion status.

Abstract: Systems, computer-implemented methods, and computer program products that can facilitate extraction of named entities from document data to support automation applications are provided. According to an embodiment, a system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise an entity extraction component that extracts, using a first machine learning process, a first data entity and a second data entity from document data indicative of a textual information. The computer executable components can further comprise a relation extraction component that determines, using a second machine learning process, a relation between the first data entity and the second data entity to generate a knowledge data graph used to execute an application associated with natural language processing for the document data.

Abstract: Systems and processes are disclosed for handling a multi-part voice command for a virtual assistant. Speech input can be received from a user that includes multiple actionable commands within a single utterance. A text string can be generated from the speech input using a speech transcription process. The text string can be parsed into multiple candidate substrings based on domain keywords, imperative verbs, predetermined substring lengths, or the like. For each candidate substring, a probability can be determined indicating whether the candidate substring corresponds to an actionable command. Such probabilities can be determined based on semantic coherence, similarity to user request templates, querying services to determine manageability, or the like. If the probabilities exceed a threshold, the user intent of each substring can be determined, processes associated with the user intents can be executed, and an acknowledgment can be provided to the user.

Abstract: An Internet of Thing (IoT) device checks user authentication using a combination of voice, image, and mobile devices.

Abstract: Methods, apparatus and articles of manufacture to identify sources of network streaming services are disclosed. An example apparatus includes a coding format identifier to identify, from a received first audio signal representing a decompressed second audio signal, an audio compression configuration used to compress a third audio signal to form the second audio signal, and a source identifier to identify a source of the second audio signal based on the identified audio compression configuration.

Abstract: Techniques performed by a data processing system for analyzing training data for a machine learning model and identifying outliers in the training data herein include obtaining training data for the model from a memory of the data processing system; analyzing the training data using a Siamese Neural Network to determine within-label similarities and cross-label similarities associated with a plurality of data elements within the training data, the within-label representing similarities between a respective data element and a first set of data elements similarly labeled in the training data, the cross-label similarities representing similarities between the respective data element and a second set of data elements dissimilarly labeled in the training data; identifying outlier data elements in the plurality of data elements based on the within-label and cross-label similarities; and processing the training data comprising the outlier data elements.

Abstract: A method and or system for processing a response message from an analytical application that includes receiving the response message; parsing the response message to facilitate selecting a semantic model to translate the response message; obtaining the semantic model to translate the response message; translating the response message using the semantic model; and converting the translated response message to non-rich text. The non-rich text can be annotated for semantic meaning that can be displayed for example on a “dumb” display that does not support rich-text formats.

Abstract: In one embodiment, a method includes accessing a machine-learning model configured to generate an encoding for an utterance by using a module to process data associated with each segment of the utterance in a series of iterations, performing operations associated with an i-th segment during an n-th iteration by the module, which include receiving an input comprising input contextual embeddings generated for the i-th segment in a preceding iteration and a memory bank storing memory vectors generated in the preceding iteration for segments preceding the i-th segment, generating attention outputs and a memory vector based on keys, values, and queries generated using the input, and generating output contextual embeddings for the i-th segment based on the attention outputs, providing the memory vector to the module for performing operations associated with the i-th segment in a next iteration, and performing speech recognition by decoding the encoding of the utterance.

Abstract: This application relates to apparatus and methods for automatically generating item information, such as item descriptions, and providing the item information to customers. For example, the embodiments may generate and provide personalized item descriptions to customers during conversational interactions in speech-based systems. In some examples, the embodiments determine entities (e.g., attributes) from item information, and apply trained machine learning processes to the extracted entities to generate textual data, such as item descriptions. For example, a computing device may apply a trained natural language processing, such as a trained transformer-based machine learning technique, to the extracted entities to generate the item descriptions. In some examples, the computing device applies post processing techniques to the generated textual data. The generated textual data may include descriptive phrases that are user friendly to customers in an e-commerce system.

Abstract: According to some embodiments, a machine learning model may include an input layer to receive an input signal as a series of frames representing handwriting data, speech data, audio data, and/or textual data. A plurality of time layers may be provided, and each time layer may comprise a uni-directional recurrent neural network processing block. A depth processing block may scan hidden states of the recurrent neural network processing block of each time layer, and the depth processing block may be associated with a first frame and receive context frame information of a sequence of one or more future frames relative to the first frame. An output layer may output a final classification as a classified posterior vector of the input signal. For example, the depth processing block may receive the context from information from an output of a time layer processing block or another depth processing block of the future frame.

Abstract: Exemplary embodiments disclose a method, a computer program product, and a computer system for generating dialogue via hashing functions. Exemplary embodiments may include detecting dialogue between one or more participants, converting the dialogue to a hashcode, and determining one or more responses to the dialogue by applying one or more models to the hashcode, wherein the one or more models correlates one or more hashcodes with the one or more responses.

Abstract: Techniques for implementing a “sticky” user ID are described. A system receives first input audio data and determines first speech processing results therefrom. The system also determines a first user ID of a user that spoke an utterance represented in the first input audio data and associates the first user ID with a device, which originated the first input audio data, for a predetermined length of time. The system determines first output data responsive to the first speech processing data and causes the device to present first output content corresponding thereto. The system then receives second input audio data and determines second speech processing results therefrom. The system also determines a time of receipt of the second input audio data is within the predetermined length of time. Based at least in part thereon, the system determined second output data responsive to the second speech processing data using the first user ID.

Abstract: A voice recognition device includes at least one position retrieving device, a directional voice receiving device, a noise suppressor, and a voice recognition processor. The position retrieving device is sequentially coupled to the directional voice receiving device, the noise suppressor, and the voice recognition processor. The position retrieving device retrieves the physical voice position of a voice source and outputs the voice position to the directional voice receiving device. The directional voice receiving device receives a voice signal generated by the voice source according to the voice position. The noise suppressor eliminates the noise of the voice signal to generate a voice recognition signal based on noise model corresponding to the voice position. The voice recognition processor receives the voice recognition signal and generates an operating signal based on the voice recognition signal.

Abstract: An automatic interpretation method performed by a correspondent terminal communicating with an utterer terminal includes receiving, by a communication unit, voice feature information about an utterer and an automatic translation result, obtained by automatically translating a voice uttered in a source language by the utterer in a target language, from the utterer terminal and performing, by a sound synthesizer, voice synthesis on the basis of the automatic translation result and the voice feature information to output a personalized synthesis voice as an automatic interpretation result. The voice feature information about the utterer includes a hidden variable including a first additional voice result and a voice feature parameter and a second additional voice feature, which are extracted from a voice of the utterer.