Abstract: A sound discriminating method comprises sensing a sound signal; changing the sensed sound signal into an electrical signal; and determining whether the electrical signal is a predetermined sound by analyzing the electrical signal.

Abstract: Aspects of the present disclosure relate to training a conversational agent. A conversational agent is initiated with a set of atomic operators, each atomic operator having one or more atomic operator rules for executing operations. A natural language input is received from a user on a conversational interface. A determination is made whether the natural language input maps to an existing script by referencing a rules engine. In response to a determination that the natural language does not map to an existing script, training is requested from the user. Training is then received from the user which maps the natural language input to a script. A mapping rule which maps the natural language input to the script is stored, and the script is executed.

Abstract: The present disclosure provide a signal processing method and apparatus. The method includes: determining a total quantity of to-be-allocated bits corresponding to a current frame; implementing primary bit allocation on to-be-processed sub-bands; performing a primary information unit quantity determining operation for each sub-band that has undergone the primary bit allocation; selecting sub-bands for secondary bit allocation from the to-be-processed sub-bands according to at least one of a sub-band characteristic of each sub-band of the to-be-processed sub-bands or the total quantity of surplus bits; implementing secondary bit allocation on the sub-bands for secondary bit allocation; and performing a secondary information unit quantity determining operation for each sub-band of the sub-bands for secondary bit allocation.

Abstract: A method, apparatus and computer program product for creating a dialog system for web content is described. Knowledge is extracted from a target web application for the dialog system. The knowledge includes an organizational structure of the target web application and domain knowledge pertinent to the target web application. A deep learning process associates the domain knowledge with the organization structure of the target application. A plurality of knowledge sources of different respective types are created from the domain knowledge and the organizational structure. Each of the knowledge sources is used for providing answers to user queries to the dialog system. As part of the invention, a semantic matcher is provided to select among the answers provided by the plurality of knowledge sources for a best answer to a user query.

Abstract: Embodiments of the present disclosure pertain to extracting data corresponding to particular data types using machine learning algorithms. In one embodiment, a method includes receiving an image in a backend system, sending the image to an optical character recognition (OCR) component, and in accordance therewith, receiving a plurality of characters recognized in the image. The character set is matched against known values to generate candidate character strings. The character set is processed by one or more machine learning algorithms to produce features. For each candidate character string, the features are then processed by a random forest model to determine a final character string.

Abstract: In one aspect, a computerized method useful for, with an ensemble of Natural Language Understanding and Processing methods converting a set of user actions into machine queries, includes the step of providing a knowledge model. The method includes the step of receiving a natural language user query; preprocesses the natural language user query for further processing as a preprocessed user query. The preprocessing includes the step of chunking a set of sentences of the natural language query into a set of smaller sentences and retaining the reference between chunks of the set of sentences. The method includes the step of, with the preprocessed user query. For each chunk of the chunked preprocessed user query the method implements the following steps.

Abstract: In one embodiment, a method includes receiving a request to generate k keywords each of which is semantically related to a particular subject, where the request includes an input n-gram representing the particular subject, accessing a table of word vector relationships, where the table includes a plurality of unique n-grams and their corresponding word vectors, and wherein each of the word vectors represents a semantic context of a corresponding n-gram as a point in a d-dimensional embedding space, looking up, using the table, a first word vector corresponding to the input n-gram, selecting k word vectors closest to the first word vector in the embedding space using the table and based on a similarity metric, identifying, for each of the selected word vectors, a corresponding n-gram by looking up the selected word vector in the table, and sending a response message including the identified n-grams.

Abstract: A digital assistant supported on computing devices such as smartphones, tablet computers, personal computers (PCs), wearable devices, game consoles, connected car devices, virtual and mixed reality devices, and the like is configured to interact with a web browser to open multiple websites within a group responsively to a single voice input at a natural language interface. The user can select websites that are included in a given group through a user interface (UI), or the computing device may be configured to automatically create or recommend groups to the user based on monitored interactions and/or user behaviors (where the monitoring is performed with notice to the user and user consent).

Abstract: The present disclosure discloses an information input method and device, and a computing apparatus. The information input method comprises receiving a voice input of a user, acquiring a recognition result on the received voice input, and enabling editing of the acquired recognition result in a text format. With the information input mechanism, according to the present invention, a user is able to choose to revise an automatic speech recognition result in a text editing format, particularly in the case where a small amount of errors occurs to the contents of speech recognition. As a result, the trouble that all contents of a voice input need to be input again is avoided, the speech recognition efficiency is increased, and the user experience is improved.

Abstract: A natural language processing system that includes an artificial intelligence (AI) engine and a tag management engine. The AI engine is configured to receive a set of audio files and to identify concepts within the set of audio files. The AI engine is further configured to determine a usage frequency for each of the identified concepts and to generate an AI-defined tag for concepts with a usage frequency that is greater than a usage frequency threshold. The tag management engine is configured to receive an audio file, identify tags linked with the audio file, to determine an access frequency for the audio file within a predetermined time period, and to adjust the activity level of the tags based on the access frequency. The tag management engine is further configured to remove tags from the set of tags with an activity level that is less than a purge threshold.

Abstract: An information processing apparatus identifies, by using an audio signal acquired by collecting a user's voice, evaluation target time that includes at least either time not including the user's voice or time during which the user is producing a meaningless utterance and produces an output appropriate to the identified evaluation target time.

Abstract: The methods, apparatus, and systems described herein assist a user with a request. The methods in part receive input from a user that includes a voice input, a gesture input, a text input, biometric information, or a combination thereof; retrieve or determine a personality type of the user based on the input; determine a distress level or an engagement level of the user; determine a set of outputs responsive to the received input; rank the outputs in the set based on the retrieved or determined personality type and the determined distress level or engagement level; deliver a ranked output to the input in a modality based on the retrieved or determined personality type and a type of device configured to deliver the ranked output to the user, wherein the device comprises a navigation system, a car, a robot, or a combination thereof; and weigh the ranked output for future interactions.

Abstract: Embodiments of the disclosure generally relate to a dialog system allowing for automatically reactivating a speech acquiring mode after the dialog system delivers a response to a user request. The reactivation parameters, such as a delay, depend on a number of predetermined factors and conversation scenarios. The embodiments further provide for a method of operating of the dialog system. An exemplary method comprises the steps of: activating a speech acquiring mode, receiving a first input of a user, deactivating the speech acquiring mode, obtaining a first response associated with the first input, delivering the first response to the user, determining that a conversation mode is activated, and, based on the determination, automatically re-activating the speech acquiring mode within a first predetermined time period after delivery of the first response to the user.

Abstract: Systems, methods, and devices related to providing content items from a collection of content items. The embodiments disclosed enhance user experience by keeping track of which content items were already played to a user, and skipping those content items automatically. The embodiments disclosed herein also can enable a user to limit playing of content items to those that have been added since the last time contents items from the collection of content items were played. If the user selects to receive additional content items after receiving the newly added items, the system can provide the older unplayed content items after the newly added content news items have been played.

Abstract: Provided are methods of dynamically and selectively providing audio and video feedbacks in response to a voice command. A method may include recognizing a voice command in a user speech received through a user device, generating at least one of audio data and video data by analyzing the voice command and associated context information, and selectively outputting the audio data and the video data through at least one of a display device and a speaker coupled to a user device based on the analysis result.

Abstract: Techniques for speech processing using a deep neural network (DNN) based acoustic model front-end are described. A new modeling approach directly models multi-channel audio data received from a microphone array using a first model (e.g., multi-channel DNN) that takes in raw signals and produces a first feature vector that may be used similarly to beamformed features generated by an acoustic beamformer. A second model (e.g., feature extraction DNN) processes the first feature vector and transforms it to a second feature vector having a lower dimensional representation. A third model (e.g., classification DNN) processes the second feature vector to perform acoustic unit classification and generate text data. These three models may be jointly optimized for speech processing (as opposed to individually optimized for signal enhancement), enabling improved performance despite a reduction in microphones and a reduction in bandwidth consumption during real-time processing.

Abstract: Example aspects of the present disclosure are directed to processing voice commands or utterances. For instance, data indicative of a voice utterance can be received. A device topology representation can be accessed. The device topology representation can define a plurality of smart devices associated with one or more structures. The device topology representation can further define a location of each of the plurality of devices within the associated structures. A transcription of the voice utterance can be determined based at least in part on the device topology representation. One or more selected devices and one or more actions to be performed by the one or more selected devices can be determined based at least in part on the determined transcription and the device topology representation.

Abstract: The lossless coding method includes selecting one of a first coding method and a second coding method, based on a range in which a quantization index of energy is represented, and coding the quantization index by using the selected coding method. The lossless decoding method includes determining a coding method of a differential quantization index of energy included in a bitstream and decoding the differential quantization index by using one of a first decoding method and a second decoding method based on a range in which a quantization index of energy is represented, in response to the determined coding method.

Abstract: A mechanism is described for facilitating automatic gain adjustment in audio systems according to one embodiment. A method of embodiments, as described herein, includes determining status of one or more of gain settings, mute settings, and boost settings associated with one or more microphones based on a configuration of a computing device including a voice-enabled device. The method may further comprise recommending adjustment of microphone gain based on the configuration and the status of one or more of the gain, mute, and boost settings, and applying the recommended adjustment of the microphone gain.

Abstract: Various embodiments relating to detecting at least one of conversation, the presence and the identity of others during presentation of digital content on a computing device. When another person is detected, one or more actions may be taken with respect to the digital content. For example, the digital content may be minimized, moved, resized or otherwise modified.