Abstract: A system and method for the management of digital assistants enabling the generation and issuance of audible user reminders in response to previous digital assistant requests issued by the user. In particular, the issued reminder is related to the operation or state of a particular system or device controlled or otherwise impacted as a consequence of the user's previous digital assistant request. Such reminders are issued after a predetermined time has elapsed after the issuance of the digital assistant request. The reminders can be persistent until the system receives an acknowledgement from the user or feedback that the state or operation of the particular impacted system or device has been altered.

Abstract: Various embodiments of the present invention relate to an electronic device for performing voice recognition using microphones selected on the basis of the operation state, and an operation method of same. According to an embodiment, the electronic device includes: one or more microphone arrays which include a plurality of microphones; at least one processor operatively connected to the microphone arrays; and at least one memory electrically connected to the processor, wherein the memory may store instructions for the processor to, at the time of execution; receive wake-up utterances, for calling designated voice services, by using a first group of microphones among the plurality of microphones when operating in a first state; operate in a second state in response to the wake-up utterances; and receive subsequent utterances using a second group of microphones among the plurality of microphones when operating in the second state. Various other embodiments are also possible.

Abstract: A first speech audio file scheduled for transmission in a first position of a transmission schedule, a first transcript of the first speech audio file, and first metadata associated with a first content item scheduled for transmission in a second position of the transmission schedule are obtained. The first metadata is compared to at least a portion of the first transcript to determine whether any of the first metadata matches the at least a portion of the first transcript. In response to determining that at least a portion of the first metadata matches at least a portion of the first transcript, linkages are created between the first speech audio file and the first content item. A graphical user interface (GUI) is generated, and displays a list of items scheduled for transmission including: a first identifier representing the first speech audio file in the first position; and a second identifier representing the first content item in the second position.

Abstract: An electronic device includes a communication module, a plurality of microphones, and a processor. The processor is configured to identify a position of a sound source based on a voice received through the plurality of microphones, identify whether the position of the sound source is included in a first zone between the electronic device and an access point that transmits and receives a communication signal with the electronic device, identify whether the voice has been uttered by a user based on a comparison between the communication signal and a preset communication signal, when the position of the sound source is included in the first zone, and determine whether to execute a command included in the voice based on the identification of whether the voice has been uttered by the user.

Abstract: Provided is an information processing apparatus and an information processing method which can adaptively switch a user interface to be used by a user to environmental information. The information processing apparatus includes an interface control unit that switches a user interface to be used by a first user at least between a first user interface using voice and a second user interface different from the first user interface on the basis of environmental information.

Abstract: Implementations set forth herein relate to an automated assistant that can operate in a transient personalization mode, and/or assist a separate automated assistant with providing output according to a transient personalization mode. The transient personalization mode can allow a guest user of an assistant enabled-device to receive personalized responses from the assistant-enabled device—despite not being signed into the assistant-enabled device. A host automated assistant of the assistant-enabled device can securely communicate with a guest user's automated assistant through a backend process. In this way, input queries from the guest user to the host automated assistant can be personalized according to the guest automated assistant—without the guest user directly engaging with their own personal device.

Abstract: An agent device includes a microphone configured to collect a speech sound inside a vehicle cabin, a speaker configured to output a speech sound inside the vehicle cabin, and a plurality of agent function elements having different functions. Each of the plurality of agent function elements generates a speech sound of an agent that speaks to an occupant of a vehicle on the basis of a meaning of the speech sound collected by the microphone, causes the speaker to output the generated speech sound of the agent. The agent function element serving as a main agent that has a dialogue with the occupant is switched on the basis of content of the speech sound collected by the microphone.

Abstract: A system for remotely and interactively controlling visual and multimedia content displayed on and rendered by a web browser using a telephony device. In particular, the system relates to receiving a voice input (e.g., dual tone multi-frequency DTMF input, spoken input, etc.) from a telephony device (e.g., a landline, a cellular telephone, or other system with telephone functionality, etc.) via a wide-area network to an intermediary computer that is configured to control the rendering of one or more web pages (or other web data) by a standard web browser.

Abstract: Techniques for responding to a user input using a device-executable intent are described. A user may speak, to a device, an input corresponding to a first natural language understanding (NLU) intent executable by a speech processing system. The speech processing system may determine a device type corresponding to the device and, based on the device type, determine a data subscription is required to execute the first NLU intent. Thereafter, the speech processing system may determine a second NLU intent that is executable by the device and that results in the output of similar content to the execution of the first NLU intent. The speech processing system then directs the device to execute the second NLU intent.

Abstract: To provide a bandwidth extension method which allows reduction of computation amount in bandwidth extension and suppression of deterioration of quality in the bandwidth to be extended. In the bandwidth extension method: a low frequency bandwidth signal is transformed into a QMF domain to generate a first low frequency QMF spectrum; pitch-shifted signals are generated by applying different shifting factors on the low frequency bandwidth signal; a high frequency QMF spectrum is generated by time-stretching the pitch-shifted signals in the QMF domain; the high frequency QMF spectrum is modified; and the modified high frequency QMF spectrum is combined with the first low frequency QMF spectrum.

Abstract: Manual human processing of documents often generates results that are subjective and include human-error. The cost and relatively slow speed of manual, human analysis makes it effectively impossible or impracticable to perform document analysis at the scale, speed, and cost desired in many industries. Accordingly, it may be advantageous to employ objective, accurate rule-based techniques to evaluate and process documents. This application discloses data processing equipment and methods specially adapted for a specific application: analysis of the breadth of documents. The processing may include context-dependent pre-processing of documents and sub-portions of the documents. The sub-portions may be analyzed based on word count and commonality of words in the respective sub-portions. The equipment and methods disclosed herein improve upon other automated techniques to provide document processing by achieving a result that quantitatively improves upon manual, human processing.

Abstract: A question answering method includes obtaining target question information; determining a candidate question and answer pair based on the target question information; calculating a confidence of answer information in the candidate question and answer pair, where the confidence is used to indicate a probability that question information in the candidate question and answer pair belongs to an answer database or an adversarial database; determining whether the confidence is less than a first preset threshold; and when the confidence is less than the first preset threshold, outputting information indicating incapable of answering.

Abstract: A method of speaker identification comprises receiving an audio signal representing speech; performing a first voice biometric process on the audio signal to attempt to identify whether the speech is the speech of an enrolled speaker; and, if the first voice biometric process makes an initial determination that the speech is the speech of an enrolled user, performing a second voice biometric process on the audio signal to attempt to identify whether the speech is the speech of the enrolled speaker. The second voice biometric process is selected to be more discriminative than the first voice biometric process.

Abstract: Disclosed is a device for controlling a plurality of voice recognition devices for determining and selecting a first voice recognition device that a user wants to use based on a point in time when the voice of the user is spoken or a place where the user spoke the voice. The device for controlling a plurality of voice recognition devices according to the present disclosure may be associated with an artificial intelligence module, a robot, an augmented reality (AR) device, a virtual reality (VR) device, a device related to 5G service, etc.

Abstract: A communication system for processing a call includes control logic and at least one machine learning model generating call classifiers from outputs of an audio signal processor and a natural language processor operated on the call. Heuristic logic transforms the call classifiers into weighted sub-metrics for the call, and aggregate normalized Gaussian logic transforms the weighted sub-metrics into a metric control that may be applied as a feedback signal to adapt the operation of the control logic. The control logic in turn may adapt the behavior of an agent, automated voice attendant, or a template utilized in a call flow.

Abstract: An alert generator in a communication system for processing a call includes at least one machine learning model generating call classifiers from outputs of an audio signal processor and a natural language processor configure to operate on the call. Heuristic logic is configured to transform the call classifiers into a plurality of weighted sub-metrics for the call, and aggregate normalized Gaussian logic is configured to transform the weighted sub-metrics into a metric control. A threshold analyzer is configured to generate an alert signal to the communication system based on the metric control meeting a condition.

Abstract: A system incorporating configurable keywords. The system can detect a keyword in audio data and execute one function for the keyword if a first application is operating, but a second function for the keyword if a second function is operating. Each keyword may be associated with multiple different functions. If a keyword is recognized during keyword detection, a function associated with that keyword is determined based on another application running on the system. Thus detection of a same keyword may result in a different function based on system context.

Abstract: An electronic device or method for adjusting a gain on a voice operated control system can include one or more processors and a memory having computer instructions. The instructions, when executed by the one or more processors causes the one or more processors to perform the operations of receiving a first microphone signal, receiving a second microphone signal, updating a slow time weighted ratio of the filtered first and second signals, and updating a fast time weighted ratio of the filtered first and second signals. The one or more processors can further perform the operations of calculating an absolute difference between the fast time weighted ratio and the slow time weighted ratio, comparing the absolute difference with a threshold, and increasing the gain when the absolute difference is greater than the threshold. Other embodiments are disclosed.

Abstract: An electronic device, a system, a method, and a program for externally adding voice control functionality to another device are disclosed. The electronic device includes memory, a processor, an audio input interface, a first communication interface, a second communication interface, and an audio encoder. First return audio data is received by the first communication interface from a sink device. An audio signal corresponding to a voice command is received by the audio input interface from a transducer. The audio signal is encoded by the audio encoder to obtain second return audio data. The second return audio data is output by the second communication interface to a source device. The second return audio data is encoded by an audio encoder of the source device to obtain control data. A function of the source device is controlled based on the control data.

Abstract: A voice conversation system for questioning a target person to encourage him/her to perform an action recommended for him/her including a response content determination unit that determines, when the questioning is performed, whether there has been a response assumed in advance from the target person in response to the questioning and ends the questioning when the response content determination unit determines that there has been a response assumed in advance from the target person, and an engagement level determination unit that calculates an engagement level, which indicates how enthusiastic the target person is in continuing a conversation, when the response content determination unit determines that there has been no response assumed in advance from the target person, and when the calculated engagement level is a first threshold or higher, questions the target person again, while when the calculated engagement level is lower than the first threshold, ends the questioning.