Abstract: A system for managing media content annotations is configured to generate annotations having a format similar to a title and tailored to a user profile. The system identifies a media content item and identifies a user entity. The system selects from among a plurality of annotations linked to the media content item and stored in metadata. For example, the system may generate more than one annotation, generate links between each annotation and user profile information, and then select among the annotations for the most appropriate annotation for a given user. The annotation may include keywords or entities that are included in, linked to, or otherwise associated with the user profile information. The system outputs, or generates for output, a display that includes a representation of the media content item and the selected annotation.

Abstract: Systems and methods for a computer-based process that determines when a viewer is likely to skip over supplemental content and adjusts supplemental content presentation to compensate. Systems of embodiments of the disclosure may utilize various inputs to determine the likelihood of skipping supplemental content, including cursor position at or near specified icons or other UI elements, as well as user actions such as gaze direction, various motions or actions, controller manipulations, and the like. Once a likelihood of skipping supplemental content is determined, various actions may be taken in response, including without limitation selection of supplemental content that conveys its intended message prior to skipping, supplemental content that can be played at increased speed, and designation of supplemental content slots as skippable or non-skippable.

Abstract: The system provides a voice command recommendation to a user to avoid a non-voice command. The system determines a command that is expected to be received, and generates a voice command recommendation that corresponds to the predicted command. The predicted command can be based on the user's behavior, a plurality of users' behavior, environmental circumstances such as a phone call ring, or a combination thereof. The system may access one or more databases to determine the predicted command. The voice command recommendation may include a displayed notification that describes the recommended voice command, and exemplary voice inputs that are recognized. The system also activates an audio interface, such as a microphone, that is configured to receive a voice input. If the system receives a recognizable voice input at the audio interface that corresponds to the recommendation, the system performs the predicted command in response to receiving the voice input.

Abstract: An exemplary automatic speech recognition (ASR) system may receive an audio input including a segment of speech. The segment of speech may be independently processed by general ASR and domain-specific ASR to generate multiple ASR results. A selection between the multiple ASR results may be performed based on respective confidence levels for the general ASR and domain-specific ASR. As incremental ASR is performed, a composite result may be generated based on general ASR and domain-specific ASR.

Abstract: Systems and methods are disclosed and described for correcting errors in ASR transcriptions. For an incorrect transcription, different words or phrases from the transcription, and/or related words or phrases, are submitted as hint words to the ASR system, and the voice query is submitted again, to determine new transcriptions. This process is repeated with different transcription terms, until a different and more proper transcription is generated. This increases the accuracy of ASR systems.

Abstract: Systems and methods are provided herein for avoiding inadvertently trigging a voice assistant with audio played through a speaker. An audio signal is captured by sampling a microphone of the voice assistant at a sampling frequency that is higher than an expected finite sampling frequency of previously recorded audio played through the speaker to generate a voice data sample. A quality metric of the generated voice data sample is calculated by determining whether the generated voice data sample comprises artifacts resulting from previous compression or approximation by the expected finite sampling frequency. Based on the calculated quality metric, it is determined whether the captured audio signal is previously recorded audio played through the speaker. Responsive to the determination that the captured audio signal is previously recorded audio played through the speaker, the voice assistant refrains from being activated.

Abstract: Systems and methods for correcting recognition errors in speech recognition systems are disclosed herein. Natural conversational variations are identified to determine whether a query intends to correct a speech recognition error or whether the query is a new command. When the query intends to correct a speech recognition error, the system identifies a location of the error and performs the correction. The corrected query can be presented to the user or be acted upon as a command for the system.

Abstract: Systems and methods detecting a spoken sentence in a speech recognition system are disclosed herein. Speech data is buffered based on an audio signal captured at a computing device operating in an active mode. The speech data is buffered irrespective of whether the speech data comprises a signature word. The buffered speech data is processed to detect a presence of the sentence comprising at least one command and a query for the computing device. Processing the buffered speech data includes detecting the signature word in the buffered speech data, and in response to detecting the signature word in the speech data, initiating detection of the sentence in the buffered speech data.

Abstract: Systems and methods are provided herein for determining one or more traits of a speaker based on voice analysis to present content item to the speaker. In one example, the method receives a voice query and determines whether the voice query matches within a first confidence threshold of a speaker identification (ID) among a plurality of speaker IDs stored in a speaker profile. In response to determining that the voice query matches to the speaker ID within the first confidence threshold, the method bypasses a trait prediction engine and retrieves a trait among the plurality of traits in the speaker profile associated with the matched speaker ID. The method further provides a content item based on the retrieved trait.

Abstract: Systems and methods are described herein for interpreting natural language search queries that account for contextual relevance of words of the search query that would ordinarily not be processed, including, for example, processing each word of the query. Each term or phrase is associated with a respective part of speech, and a frequency of occurrence of a combination of adjacent terms or phrases public domain is determined. A relevance of each term is then determined based on its respective type of term and frequency of occurrence in the public domain. The natural language search query is then interpreted based on the importance or relevance of each term.

Abstract: Systems and methods for modifying a notification received on a user device are disclosed. The method comprises receiving from at least one application on a user device a notification, identifying temporal information associated with the notification, detecting based on the temporal information a calendar-related event from a user profile, and modifying the temporal information associated with the notification to refer to the calendar-related event.

Abstract: A transcription of a query for content discovery is generated, and a context of the query is identified, as well as a first plurality of candidate entities to which the query refers. A search is performed based on the context of the query and the first plurality of candidate entities, and results are generated for output. A transcription of a second voice query is then generated, and it is determined whether the second transcription includes a trigger term indicating a corrective query. If so, the context of the first query is retrieved. A second term of the second query similar to a term of the first query is identified, and a second plurality of candidate entities to which the second term refers is determined. A second search is performed based on the second plurality of candidates and the context, and new search results are generated for output.

Abstract: Systems and methods are provided herein for training a natural language understanding model. A natural language query, e.g., a first natural language query, is received. A first natural language understanding model is used to process the natural language query. A confidence level, e.g., a first confidence level, of the understanding of the natural language query is determined. In response to the confidence level being below a confidence level threshold, the natural language query is reprocessed using a reprocessing module, e.g., after providing a response to the natural language query. The first natural language model is updated based on the reprocessing of the natural language query.

Abstract: Systems and methods are provided for responding to a natural language query, e.g., a first natural language query. A first natural language understanding model is used to process the natural language query. A confidence level, e.g., a first confidence level, of the understanding of the natural language query is determined. In response to the confidence level being below a confidence level threshold, the natural language query is reprocessed using a reprocessing module. A response to the first natural language query is generated based on the processing of the natural language query by the first natural language understanding model and the reprocessing of the natural language query.

Abstract: The system provides a voice command recommendation to a user to avoid a non-voice command. The system determines a command that is expected to be received, and generates a voice command recommendation that corresponds to the predicted command. The predicted command can be based on the user's behavior, a plurality of users' behavior, environmental circumstances such as a phone call ring, or a combination thereof. The system may access one or more databases to determine the predicted command. The voice command recommendation may include a displayed notification that describes the recommended voice command, and exemplary voice inputs that are recognized. The system also activates an audio interface, such as a microphone, that is configured to receive a voice input. If the system receives a recognizable voice input at the audio interface that corresponds to the recommendation, the system performs the predicted command in response to receiving the voice input.

Abstract: Systems and methods are described herein for interpreting natural language search queries that account for contextual relevance of words of the search query that would ordinarily not be processed, including, for example, processing each word of the query. Each term is associated with a respective part of speech, and a frequency of occurrence of each term in content metadata is determined. A relevance of each term is then determined based on its respective part of speech and frequency. The natural language search query is then interpreted based on the importance or relevance of each term.

Abstract: An exemplary automatic speech recognition (ASR) system may receive an audio input including a segment of speech. The segment of speech may be independently processed by general ASR and domain-specific ASR to generate multiple ASR results. A selection between the multiple ASR results may be performed based on respective confidence levels for the general ASR and domain-specific ASR. As incremental ASR is performed, a composite result may be generated based on general ASR and domain-specific ASR.

Abstract: Systems and methods for determining the answers to user queries regarding content, and presenting options for the user to view the answers. Searches are conducted to determine the answer to a query and, once the answer is found, various options are presented to the user, allowing the user the choice to view the answer in multiple different ways.

Abstract: Systems and methods for determining hint words that improve the accuracy of automated speech recognition (ASR) systems. Hint words are determined in the context of a user issuing voice commands in connection with a voice interface system. Terms are initially taken from most frequently occurring terms in operation of a voice interface system. For example, most frequently occurring terms that arise in electronic search queries or received commands are selected. Certain of these terms are selected as hint words, and the selected hint words are then transmitted to an ASR system to assist in translation of speech to text.

Abstract: Systems and methods are described for managing presentation of content. A determination may be made, based on user information related to the current media asset, whether the presentation of the current media asset is not likely to be of interest to the user. In response to determining that the presentation of the current media asset is not likely to be of interest to the user, a selectable option may be presented to skip current content (e.g., advertisement content, or another currently presented media asset) and start presentation of a next media asset (e.g., a selected media asset, or a media asset similar to the current media asset).