Abstract: Methods and systems are presented for providing automated online chat assistance in an online chat session. One or more utterances transmitted from a user device of a user via the online chat session are obtained. The one or more utterances are provided to a first prediction model to predict an intent of a user. If it is determined that the first prediction model is unable to predict the intent of the user based on the one or more utterances, the one or more utterances are provided to a second prediction model. After predicting the intent of the user by the second prediction model, the intent is used by a chat robot to provide a dialogue with the user via the online chat session. The one or more utterances and the predicted intent are used to re-train the first prediction model.

Abstract: The disclosure is directed to various ways of improving the functioning of computer systems, information networks, data stores, search engine systems and methods, and other advantages. Among other things, provided herein are methods, systems, components, processes, modules, blocks, circuits, sub-systems, articles, and other elements (collectively referred to in some cases as the “platform” or the “system”) that collectively enable, in one or more datastores (e.g., where each datastore may include one or more databases) and systems. A system and method for providing conversation intelligence services may include pre-processing, transcribing, and post-processing. A conversation recording may be pre-processed generating a conversation record (e.g., conversation object). The pre-processed conversation recording may be transcribed into a transcript.

Abstract: An improved system and method of defining speeding severity levels, receiving over a wireless network from a vehicle device speed data and metadata, determining associated speeding severity levels, based at least in part on the determined associated speeding severity levels, determining whether an alert communication is to be transmitted the vehicle device, and in response to determining that an alert communication is to be transmitted the vehicle device, transmitting the alert communication via a wireless interface to the vehicle device, the alert communication configured to cause the vehicle device to provide a corresponding in-vehicle alert.

Abstract: Various embodiments of the present disclosure provide automated data compliance techniques for complex access controlled datasets subject to a plurality of data access constraints. Some of the techniques may include generating, using one or more natural language models, entity-relationship data for an access controlled dataset and generating a knowledge graph based on the entity-relationship data. The knowledge graph includes a plurality of vertices connected by a plurality of edges that may be traversed to identify a data access condition indicative of a data access violation or a data coverage violation. Some of the techniques may include generating, using the knowledge graph, a natural language condition description based on the data access condition and providing a condition alert indicative of the natural language condition description.

Abstract: Techniques for joining a device of a third user to a communication between a device of a first user and a device of a second user are described herein. For instance, two or more users may utilize respective computing devices to engage in a telephone call, a video call, an instant-messaging session, or any other type of communication in which the users communicate with each other audibly and/or visually. In some instances, a first user of the two users may issue a voice command requesting to join a device of a third user to the communication. One or more computing devices may recognize this voice command and may attempt to join a device of a third user to the communication.

Abstract: The subject invention pertains to methods and systems for classifying leukocytes using artificial intelligence called AIRFIHA (artificial-intelligence enabled reagent-free imaging hematology analyzer) that can accurately classify subpopulations of leukocytes in a label-free manner. AIRFIHA can not only subtype lymphocytes into B and T cell but is capable of sorting different types of T cells subtypes. AIRFIHA is realized through training a two-step neural network using label-free images of separated leukocytes acquired from a custom-built quantitative phase microscope. Owing to its easy operation, low cost, and strong discerning capability of complex leukocyte subpopulations, AIRFIHA is clinically translatable and can also be deployed in resource-limited settings.

Abstract: A system for the control of a medical implant in a mammal body is provided. The system comprises a first and a second part being adapted for communication with each other, in which system: the first part is adapted for implantation in the mammal body for the control of and communication with the medical implant, and the second part is adapted to be worn on the outside of the mammal body and adapted to receive control commands from a user and to transmit these commands to the first part.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for hotword detection on multiple devices are disclosed. In one aspect, a method includes the actions of receiving, by a first computing device, audio data that corresponds to an utterance. The actions further include determining a first value corresponding to a likelihood that the utterance includes a hotword. The actions further include receiving a second value corresponding to a likelihood that the utterance includes the hotword, the second value being determined by a second computing device. The actions further include comparing the first value and the second value. The actions further include based on comparing the first value to the second value, initiating speech recognition processing on the audio data.

Abstract: An electronic control device of an avionics system for implementation of a critical avionics function, comprising: a module for receiving a voice instruction signal; a speech recognition module configured to transform the voice signal into a textual transcript; a processing module configured to associate the textual transcript with at least one action to be performed; a monitoring system comprising: a control module configured to check whether the textual transcript and/or the action to be performed is consistent if and only if: a) the textual transcript and/or the action to be performed is consistent with the expected syntax, b) the textual transcript and/or the action to be performed is consistent with the expected lexical field, and c) the textual transcript and/or the action to be performed is consistent with the current context, a module for generating an associated command only if no inconsistencies are detected.

Abstract: A system configured to perform style-aware listener animation. By representing different listening styles (e.g., facial expressions) using an embedding space, a single model can be trained to generate unique facial animations for a number of distinct listeners. Thus, individual listening styles can be associated with a listener identifier, enabling the system to (i) animate a plurality of different listeners with unique nonverbal behavior and/or (ii) select a particular listener identifier or desired type of listener style with which to animate. This enables the model to be generalized to new listeners to generate additional listener facial responses without needing training data for each new listener. The model may process a listener representation style or listener identifier, along with input data corresponding to a speaker talking, to generate unique facial animation responsive to the speech.

Abstract: A system and method for temporarily disabling keyword detection to avoid detection of machine-generated keywords. A local device may operate two keyword detectors. The first keyword detector operates on input audio data received by a microphone to capture keywords uttered by a user. In these instances, the keyword may be detected by the first detector and the audio data may be indicated for speech processing. The system may determine output audio data responsive to the input audio data. The local device may process the output audio data to determine that it also includes the keyword. The device may then disable the first keyword detector while the output audio data is played back by an audio speaker of the local device. Thus the local device may avoid detection of a keyword originating from the output audio. The first keyword detector may be reactivated after a time interval during which the keyword might be detectable in the output audio.

Abstract: The present disclosure relates to chatbot systems, and more particularly, to batching techniques for handling unbalanced training data when training a model such that bias is removed from the trained machine learning model when performing inference. In an embodiment, a plurality of raw utterances is obtained. A bias eliminating distribution is determined and a subset of the plurality of raw utterances is batched according to the bias-reducing distribution. The resulting unbiased training data may be input into a prediction model for training the prediction model. The trained prediction model may be obtained and utilized to predict unbiased results from new inputs received by the trained prediction model.

Abstract: Implementations are directed to determining an audio delay, of a computing device, by causing an audio data stream to be transmitted to the computing device via a wireless communication channel. The computing device causes audio output generated using the audio data stream to be rendered via speaker(s). The rendered audio output is captured via microphone(s), and the audio delay determined by comparing the captured audio output with the audio data stream. A delay audio segment can be appended to an additional audio data stream transmitted to the computing device, where the length of the delay audio segment is determined using the audio delay. A noise reduction technique can additionally or alternatively be adapted based on the audio delay. Implementations are additionally or alternatively directed to determining if an audio data stream transmitted to a computing device for rendering through speaker(s) driven by the computing device—is actually being rendered.

Abstract: A voice control method and apparatus for a device, a storage medium, and an electronic apparatus are provided. The method includes: acquiring a first voice feature of first voice data collected by a cleaning device, where the first voice data correspond to a first wake-up instruction sent by a use object, and the first wake-up instruction is used to wake up at least one of the cleaning device and a base station; acquiring a second voice feature of second voice data collected by the base station, where the second voice data correspond to the first wake-up instruction; and selecting a first device to be woken up from the cleaning device and the base station according to the first and second voice features, and waking up the first device, where the first device in a wake-up state is configured to respond to a voice instruction sent by the use object.

Abstract: A server is disclosed. The disclosed server includes a communication device for performing communication with a home appliance and a terminal device, a memory for storing state information of the home appliance and operation pattern information obtained by analyzing an operation pattern of the home appliance, and a processor for, when a query for the home appliance is received from the terminal device, generating response information in response to the received query, and controlling the communication device to transmit the generated response information to the terminal device, wherein the processor extracts a keyword included in the received query, checks at least one state item corresponding to the extracted keyword, and generates response information by using information corresponding to at least one state item in the state information and the operation pattern information.

Abstract: A method is provided. The method is executed by an autocomplete prediction engine implemented as a computer program within a computing environment. The autocomplete prediction engine executes automated communication mining on a communication. The method includes processing the communication to extract intents and entities related to each intent. The method includes providing the intents and the entities into forms using a language model to provide a conversational or natural language understanding of the communication.

Abstract: An information processor including: an operation control unit that controls a motion of an autonomous mobile body acting on the basis of recognition processing, in a case where a target sound that is a target voice for voice recognition processing is detected, the operation control unit moving the autonomous mobile body to a position, around an approach target, where an input level of a non-target sound that is not the target voice becomes lower, the approach target being determined on the basis of the target sound.

Abstract: An interactive method and an interactive device for the meeting minute, an apparatus and a medium are provided. The method includes receiving an interactive triggering operation of a user for the meeting minute in a meeting minute display interface, where the meeting minute display interface displays a multimedia, a meeting subtitle of the multimedia and the meeting minute; and playing the multimedia based on an associated time period of the meeting minute, and distinctively displaying an associated subtitle of the meeting minute in the meeting subtitle. According to the above technical solution, the multimedia can be associated with content related to the meeting subtitle through the interactive triggering operation of the user for the meeting minute in the meeting minute display interface, to improve interactive experience effect of the user.

Abstract: A system and a method for controlling a lamp of a vehicle are provided, The system includes a voice recognition device to recognize a voice signal from a user, a voice analyzing device to analyze context of the recognized voice signal to determine a lamp device to be controlled, and to determine a control intent of the user with respect to the determined lamp device, and a controller to control the lamp device, based on the determined control intent of the user.

Abstract: A voice wakeup method includes receiving a plurality of voice wakeup messages sent by a plurality of electronic devices, where each voice wakeup message includes a distance and a wakeup energy value; determining, based on distances and wakeup energy values in the plurality of voice wakeup messages from the plurality of electronic devices, whether energy attenuation of the wakeup word emitted by the sound source complies with an attenuation law of sound energy radiated by a point source; and when determining that the energy attenuation of the wakeup word emitted by the sound source does not comply with the attenuation law of the sound energy radiated by the point source, sending a wakeup forbidding instruction to the plurality of electronic devices.

Abstract: An electronic device includes a speaker, a sensor, a communication circuit, a processor, and a memory to store instructions. The instructions, when executed by the processor, cause a wireless audio device to, while outputting a signal for reducing an external sound through the speaker, identify, using the communication circuit, an external electronic device, identify, using the sensor, a conversation responsive to a location of the external electronic device satisfying a specified condition, responsive to identifying the conversation, stop an output of the signal for reducing the external sound for a first period of time, and responsive to identifying a specified keyword included in the conversation, prolong stopping the output of the signal for reducing the external sound for a second period of time.

Abstract: Systems and methods are described herein for enabling, on a local device, a voice control system that limits the amount of data needed to be transmitted to a remote server. A data structure is built at the local device to support a local speech-to-text model by receiving a query and transmitting, to a remote server over a communication network, a request for a speech-to-text transcription of the query. The transcription is received from the remote server and stored in the data structure at the local device in association with an audio clip of the query. Metadata describing the query is used to train the local speech-to-text model to recognize future instances of the query.

Abstract: A system includes a gimbal, a shotgun microphone coupled to the gimbal, a camera, and at least one processor. The at least one processor is configured to receive data indicative of an image or video feed from the camera. The at least one processor is also configured to determine, based on the data indicative of the image or video feed, a primary human speaker among a group of humans and a location of the primary human speaker. The at least one processor is also configured to control the gimbal to point the shotgun microphone at the location of the primary human speaker.

Abstract: A set of alternative vocal input styles for specifying a parameter of a dialog-driven application is determined. During execution of the application, an audio prompt requesting input in one of the styles is presented. A value of the parameter is determined by applying a collection of analysis tools to vocal input obtained after the prompt is presented. A task of the application is initiated using the value.

Abstract: Implementations are directed to determining an audio delay, of a computing device, by causing an audio data stream to be transmitted to the computing device via a wireless communication channel. The computing device causes audio output generated using the audio data stream to be rendered via speaker(s). The rendered audio output is captured via microphone(s), and the audio delay determined by comparing the captured audio output with the audio data stream. A delay audio segment can be appended to an additional audio data stream transmitted to the computing device, where the length of the delay audio segment is determined using the audio delay. A noise reduction technique can additionally or alternatively be adapted based on the audio delay. Implementations are additionally or alternatively directed to determining if an audio data stream transmitted to a computing device for rendering through speaker(s) driven by the computing device—is actually being rendered.

Abstract: Techniques are described for training and/or utilizing sub-agent machine learning models to generate candidate dialog responses. In various implementations, a user-facing dialog agent (202, 302), or another component on its behalf, selects one of the candidate responses which is closest to user defined global priority objectives (318). Global priority objectives can include values (306) for a variety of dialog features such as emotion, confusion, objective-relatedness, personality, verbosity, etc. In various implementations, each machine learning model includes an encoder portion and a decoder portion. Each encoder portion and decoder portion can be a recurrent neural network (RNN) model, such as a RNN model that includes at least one memory layer, such as a long short-term memory (LSTM) layer.

Abstract: A method for vehicle lane changing control, a device, a storage medium and a program product, relating to the fields of autonomous driving, intelligent transportation, big data, cloud computing, etc. in computer technology. A specific implementation solution is: during a process of vehicle driving, when a distance between the vehicle and a front bifurcation is less than or equal to a longest lane changing operation distance, according to map data and traffic data of a road section ahead, it is determined whether a current position is a preferred position to start performing the lane changing operation, so as to determine whether performing the lane changing operation at the current position; and if determined to perform the lane changing operation at the current position, lane changing operation execution information will be issued to enable the vehicle to start performing the lane changing operation at the current position.

Abstract: Methods and systems are provided for displaying transcriptions of radio communication transcription for an aircraft. The method comprises capturing audio signals of radio communication traffic to and from the aircraft. The captured audio signals are preprocessed to divide the signals into independent spoken utterances. Each spoken utterance is transcribed using a speech recognition decoder that utilizes an air traffic control (ATC) speech recognition model and classification data is extracted from the transcription of each spoken utterance. The transcription of each spoken utterance is logged with reference to the classification data and a textual display of the transcription is provided to a crew member of the aircraft.

Abstract: An electronic device of the present disclosure comprises: a communication unit; a memory; and a processor for: detecting a voice section in an audio signal acquired by the electronic device; identifying whether a wake-up word stored in the memory exists in a user voice included in the detected voice section; when it is identified that the wake-up word exists in the user voice, transmitting, via the communication unit, the user voice to a server for providing a voice recognition service; and when response information for the user voice is received from the server, providing a response to the user voice on the basis of the received response information, wherein the processor identifies that the wake-up word exists in the user voice, when a part of the user voice matches the wake-up word. In particular, a method for acquiring a natural language for providing a response may use an artificial intelligence model learned according to at least one of machine learning, a neural network, and a deep learning algorithm.

Abstract: A voice control system for ophthalmologic laser treatment systems sets parameters for delivering laser energy based on voice commands and prevents potentially harmful parameters due to operator mistakes and misunderstood voice commands by providing incremental parameter adjustment and restricting the amount by which the parameters can be adjusted for each executed voice command. Valid voice commands include indications of which parameter to set, a value for the parameter, and whether to increase or decrease the value of the parameter. In one example, parameter values can only be increased or decreased by a certain percentage with respect to the current value. In another example, the parameters are adjusted by selecting the next highest or lowest value with respect to the current parameter value from a predetermined sequence of possible values for particular parameters. Voice control functionality can also be deactivated under certain conditions such as when it is determined that a parameter was not set.

Abstract: A communications system for a restaurant includes multiple field communications units, a portable communications device, and processing circuitry. The field communications units include a speaker and a microphone. The portable communications device is transportable by an order taker of the restaurant. The processing circuitry obtains a user input to transition the portable communications device between multiple different channels. The multiple different channels correspond to one of the multiple field communications units. The processing circuitry transitions the portable communications device to one of the multiple different channels according to the user input. The processing circuitry facilitates end-to-end bi-directional audio communication between the order taker and a customer by operating the portable communications device and the one of the multiple field communications units to exchange audio data over the one of the multiple different channels.

Abstract: Techniques for capturing spoken user inputs while a device is prevented from capturing such spoken user inputs are described. When a first device becomes incapable of capturing spoken user inputs intended for a system, a second device, for capturing such spoken user inputs, may be identified. The second device may be identified based on the second device being connected to a same vehicle computing system as the first device. The second device may be enabled to capture spoken user inputs, intended for the system, until the first device is again able to capture such spoken user inputs.

Abstract: Disclosed is a system for determining sequences of operations that will automatically execute one or more tasks specified by a user. In some embodiments, the sequences of operations are based on operations that have been previously performed by users and recorded by the system. The system interprets an intention of a user based on analysis of terms used by the user to indicate a request. The system generates a sequence of operations, executable by an operating system associated with a client device that will perform one or more tasks specified or implied by the request of the user.

Abstract: An apparatus and a method for controlling a vehicle sound are provided. The apparatus a detection device that detects driving information and drive mode setting information and a processing device electrically connected with the detection device. The processing device determines an emotional state of a driver based on at least one of the driving information or the drive mode setting information, determines a sound concept depending on the emotional state of the driver, and controls a vehicle sound depending on the sound concept.

Abstract: Techniques for altering audio being output by a voice-controlled device, or another device, to enable more accurate automatic speech recognition (ASR) by the voice-controlled device. For instance, a voice-controlled device may output audio within an environment using a speaker of the device. While outputting the audio, a microphone of the device may capture sound within the environment and may generate an audio signal based on the captured sound. The device may then analyze the audio signal to identify speech of a user within the signal, with the speech indicating that the user is going to provide a subsequent command to the device. Thereafter, the device may alter the output of the audio (e.g., attenuate the audio, pause the audio, switch from stereo to mono, etc.) to facilitate speech recognition of the user's subsequent command.

Abstract: Systems and methods are provided for training of an Automatic Speech Recognition (ASR) model during runtime of a transcription system, the system includes a background processor configured to operate with the transcription system to display a speech-to-text sample of an audio segment of a cockpit communication with an identifier which is converted using an ASR model wherein the background processor receives a response by a user during runtime of the transcription system and display of the speech-to-text sample and causes a change to the identifier to either a positive or negative attribute upon a determination of the correctness of a conversion process of the speech-to-text sample using the ASR model by review of a display of the content of the speech-to-text sample; and to train the ASR model based on information associated with the content of the speech-to-text sample in accordance with the response by the user.

Abstract: A voice recognition system in an aircraft is provided. The voice recognition system includes: a voice recognition controller configured to receive a voice command from a flight crew member voice interface, convert the voice command to an avionics system command, and forward the avionics system command to an avionics system for execution. The voice recognition system further includes an application controller configured to: receive the avionics system command that has been converted from the voice command; determine whether the avionics system command can be performed; cause the avionics system command to be performed when it is determined that the avionics system command can be performed; and when it is determined that the avionics system command cannot be performed, generate a message that provides a reason why the avionics system command cannot be performed and cause the message to be displayed on a display device and/or annunciated on an aural device.

Abstract: For example, a BT audio device may be configured to, during a passphrase-detection mode, monitor an audio input of the BT audio device to detect whether the audio input includes a voice signal, the passphrase-detection mode configured for detection of a predefined user passphrase to indicate a voice command to be provided from a user of the BT audio device; based on a determination that the audio input does not include the voice signal, transmit one or more null-data packets to a BT device over a BT wireless communication link between the BT audio device and the BT device; and, based on a determination that the audio input includes the voice signal, transmit one or more data packets to the BT device over the BT wireless communication link, wherein a payload of the one or more data packets includes audio data based on the audio input.

Abstract: A method for generating query response comprising receiving, by a processor, an input query from a first user; performing, by the processor, query parsing on the input query to generate a parsed query; determining, by the processor, a query type associated with the parsed query; for the query type being determined as initial query, performing: generating, by the processor, a first follow-up query to the input query based on the parsed query, and generating, by the processor, responses to the parsed query and the first follow-up query; and for the query type being determined as follow-up query to a previous query entered by a second user, performing: performing, by the processor, learning of a query sequence from the previous query to the input query, and generating, by the processor, responses to the parsed query and a second follow-up query.

Abstract: Provided is a system for feedback-improved automatic solving of production facility related tasks, including: an input interface being adapted to receive production facility related data; a semantic data enhancement module being adapted to generate semantically enhanced data based on the production facility related data; a semantic-based reasoning module being adapted to automatically provide an explainable artificial intelligence model, using the semantically enhanced data, wherein the artificial intelligence model relates to a predetermined production facility related task; a user interaction interface being adapted to output explanation data regarding the explainable artificial intelligence model to a user; and a feedback module adapted to receive feedback from the user in response to the outputted explanation data and adapted to adjust the semantic-based reasoning module and/or the user interaction interface based on the received feedback.

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: A remote control for generating output signals apt at controlling one or more electronic devices includes a sound transducer, a speech recognition unit for recognizing voice commands, a memory for storing information relative to available content of the one or more electronic devices and a control signal generating and receiving unit for generating control signals corresponding to the voice commands for controlling the one or more electronic devices.

Abstract: Example techniques involve a control hierarchy for a “smart” home having smart appliances and related devices, such as wireless illumination devices, home-automation devices (e.g., thermostats, door locks, etc.), and audio playback devices, among others. An example home includes various rooms in which smart devices might be located. Under the example control hierarchy described herein and referred to as “home graph,” a name of a room (e.g., “Kitchen”) may represent a smart device (or smart devices) within that room. In other words, from the perspective of a user, the smart devices within a room are that room. This hierarchy permits a user to refer to a smart device within a given room by way of the name of the room when controlling smart devices within the home using a voice user interface (VUI) or graphical user interface (GUI).

Abstract: A system may include a memory and a processor in communication with the memory. The processor may be configured to perform operations. The operations may include generating a content page and selecting an annotation container for the content page. The operations may include configuring the annotation container and inputting content into the annotation container. The operations may include submitting at least one attribute to the annotation container to associate the attribute with the content. The operations may include including the content and the at least one attribute in container data and extracting a model from the container data. The operations may include importing the model into a dialog skill and embedding the dialog skill into a user interface.

Abstract: Systems, methods, and storage media for performing actions in response to a determined spoken command of a user are disclosed.

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: A voice wakeup method and a device are provided. The method includes: receiving, by a third-party device, wakeup messages sent by at least two electronic devices, where the wakeup message includes wakeup keyword energy information used to indicate a wakeup keyword energy value, and the wakeup keyword energy value; normalizing wakeup keyword energy values based on ambient sound energy and/or sound collection capabilities of devices, to obtain at least two normalized wakeup keyword energy values; and based on the at least two normalized wakeup keyword energy values, sending a wakeup permission instruction to a first electronic device in the at least two electronic devices, and sending a wakeup prohibition instruction to another electronic device other than the first electronic device, where a normalized wakeup keyword energy value of the first electronic device is a maximum value. This helps improve accuracy of waking up a device nearby in a multi-device scenario.

Abstract: Disclosed are a vehicle control method and an intelligent device for controlling a vehicle. A vehicle control method according to an embodiment of the present disclosure comprises: acquiring a user request; updating the user request so as to be recognized by an application providing a service related to the user request, and providing the updated user request to the application; and providing the service through the application. Accordingly, the present invention can accurately and promptly recognize words in various forms, included in a user request, thereby providing a comparatively accurate service in accordance with the user request. One or more of an autonomous driving vehicle and an intelligent computing device of the present disclosure may be linked to an artificial intelligence module, a drone (unmanned aerial vehicle, UAV), a robot, an augmented reality (AR) device, a virtual reality (VR) device, a device related to 5G services, and the like.

Abstract: Embodiments of the present disclosure provide an information input method, a system of a cloud input method and a client, where the method includes: generating an input method startup instruction according to received indication information, where the indication information is generated when a focus of an information input box is acquired; sending the input method startup instruction, where the input method startup instruction is used to startup a local input method; receiving text content, where the text content is content input by using the local input method; and submitting the text content to the information input box for display. According to the embodiments of the present disclosure, the local input method of the client can be called through the input method startup instruction, and a user uses the local input method commonly used by the client to input information, which greatly improves convenience of operations.

Abstract: Domain-specific parameters may be used for tuning speech processing. A pre-trained transformer-based language model may train domain-specific parameters using domain-specific unlabeled text data. This domain-specific parameters can then be appended to candidate texts produced by a speech model on received speech data and input to the transformer-based language model to score the candidate texts. The scores of the candidate texts determined using the pre-trained transformer-based language model can then be used to select a candidate text for further speech processing.