Abstract: Systems, methods, and devices disclosed herein can capture an audio recording of an utterance, generate a transcription based on the audio recording, and generate a score for a section of the transcription that reflects a level of confidence that at least one word in the section was correctly transcribed. Content of the section is rendered in a field on a display. Also, a timeline for the audio recording is rendered. If the score does not satisfy a condition, the fill scheme applied to a segment of the timeline that maps to the section may differ from a fill scheme that is applied to the remainder of the timeline. An additional audio recording is then captured and transcribed. An additional timeline is rendered for the additional audio recording alongside the timeline and is aligned with the section. The transcription of the additional audio recording is used to replace the section.

Abstract: In some implementations, the techniques may include receiving event data as a stream of event instances. Each received event instance is associated with an entity and a capacity change. In addition, the techniques may include identifying an event time for each received event instance. The techniques may include sorting the received event instances into sets of instances. Each set of instances can be associated with a respective entity. Moreover, the techniques may include segmenting each set of instances into subsets of instances based on the event time for each event instance of the set of instances. Each of the plurality of subsets of instances can correspond to a time period. Also, the techniques may include storing each segmented set of instances as stored event data. Further, the techniques may include performing one or more operations with respect to the stored event data.

Abstract: Systems and methods are described herein for generating a playlist for a simultaneous presentation of a plurality of media assets. The system retrieves a user preference associated with a user profile and receives a selection of a first media asset and a second media asset from the plurality of media assets for presentation on a user device. The system parses the respective audio streams of the first media asset and the second media asset to identify one or more preferred audio segments based on the user preference and generates the playlist of the identified one or more preferred audio segments. Based on a generated audio playlist, the system generates, for presentation on the user device, the video stream for each of the first media asset and the second media asset and the playlist of the identified one or more preferred audio segments.

Abstract: Systems and methods for recording a meeting using a retroactive record feature. The present technology provides for improved systems and methods for providing a recording of a virtual meeting, where a selection to initiate the recording from the beginning or an earlier time in the meeting from a current time may be received after the virtual meeting has started. The system may process received meeting content streams to generate a plurality of data segments that may collectively form a meeting recording. Each data segment, for example, may include meeting content associated with a particular user/attendee and associated with a timestamp and/or time duration. In some examples, the plurality of data segments may be stored on a blockchain, which may provide an immutable meeting record that may be concatenated together and made available for playback based on a selection to record the meeting and consent given by the users/attendees.

Abstract: The present invention relates to methods for searching for two-dimensional or three-dimensional objects. More particularly, the present invention relates to searching for two-dimensional or three-dimensional objects in a collection by using a multi-modal query of image and/or tag data. Aspects and/or embodiments seek to provide a method of searching for digital objects using any combination of images, three-dimensional shapes and text by embedding the vector representations for these multiple modes in the same space. Aspects and/or embodiments can be easily extensible to any other type of modality, making it more general.

Abstract: A system includes a computer including a processor and a memory. The memory includes instructions such that the processor is programmed to receive an audio input representing a percussion performed by a user and classify, at a trained neural network, the audio input as a particular musical type.

Abstract: Methods, apparatuses and systems directed to pattern identification and pattern recognition. In some particular implementations, the invention provides a flexible pattern recognition platform including pattern recognition engines that can be dynamically adjusted to implement specific pattern recognition configurations for individual pattern recognition applications. In some implementations, the present invention also provides for a partition configuration where knowledge elements can be grouped and pattern recognition operations can be individually configured and arranged to allow for multi-level pattern recognition schemes.

Abstract: A method including receiving a request to access a meeting record from a user is provided. The meeting record may indicate at least one meeting participant, an audio/video recording and a presentation from one of the participants in the meeting. The method includes verifying an access privilege of the user for the meeting record, providing the meeting record to the user, for playback of a selected portion, and providing, in the meeting record, a selecting tool to the user, for playing the selected portion, wherein the selecting tool is configured to playback the selected portion for one of multiple participants in the meeting.

Abstract: A framework for efficiently importing content into a speech-controlled system in a manner that makes the content easily accessible using voice commands. A speech-controlled system that can be controlled using a variety of commands, including a command to retrieve audio content, can be configured using a framework of content organization that allows new content to be ingested using the framework, thus making the new content accessible to users of the system without manually adjusting the system to recognize when incoming commands call for the new content. The framework can include configured content demarcations (such as information demarcations that divide content into articles, or other sized portions), labels for those demarcations (such as topic descriptors or the like), etc.

Abstract: Embodiments for creating compact example subsets for intent classification in a conversational system are provided. A set of content used for training an intent classifier is received from a conversational corpus. Entries within the set of content are separated into a first subset and a second subset, and a cross-validation operation is performed on the first and second subsets to identify a correctly labeled portion and an incorrectly labeled portion of the set of content. A reduced content used for performing a final training of the intent classifier is formed by combining a first number of the entries from the correctly labeled portion and a second number of the entries from the incorrectly labeled portion of the set of content.

Abstract: A method including receiving a request to access a meeting record from a user is provided. The meeting record includes an identification for at least one participant in a meeting, an audio recording for the at least one participant, a video recording for the at least one participant, and a presentation from one of the participants in the meeting. The method includes verifying an access privilege of the user for the meeting record, providing the meeting record to the user, for playback of a selected portion, and providing, in the meeting record, a selecting tool to the user, for playing the selected portion, wherein the selecting tool is configured to playback the selected portion for one of multiple participants in the meeting.

Abstract: Various embodiments of an apparatus, methods, systems and computer program products described herein are directed to a Topic Engine. The Topic Engine captures a plurality of content identifier sequences. Each respective sequence represents an order at which a corresponding user account accessed content. The Topic Engine generates a plurality of clusters. Each cluster is associated with respective content identifiers appearing within a proximity to each other across the plurality of content identifier sequences of different user accounts. The Topic Engine obtains one or more sample content identifiers from at least one cluster via sampling the cluster. The Topic Engine extracts keywords from content represented by the one or more sampled content identifiers. The Topic Engine identifies a topic for the cluster based on the one or more extracted keywords.

Abstract: An equalizer and a method of controlling same are provided. The equalizer includes a memory storing an EQ value set for a plurality of music attributes and storing a general-purpose EQ value; and a processor configured to: obtain an input music signal; calculate a plurality of probability values for the plurality of music attributes by analyzing attributes of the input music signal based on a convolutional neural network; calculate a moderate index between the plurality of probability values; generate an EQ value based on the plurality of probability values and the moderate index; and perform equalizing by applying the generated EQ value to the input music signal.

Abstract: Method and apparatus for obtaining audio corresponding to a plurality of images, based on semantic information and the emotion information of the plurality of images.

Abstract: In an information recording system, a sound processing unit generates a conversion candidate word in a process of converting sound information into text information. A recording unit records the text information and the conversion candidate word on a recording medium such that the text information and the conversion candidate word are associated with each other. A search unit performs a search based on a keyword and extracts a word matching the keyword from words within the text information and the conversion candidate word. A reading unit reads the text information including the word matching the keyword from the recording medium. A display unit displays the text information such that a part corresponding to the word matching the keyword and a part other than the corresponding part are able to be distinguished.

Abstract: The present invention is intended to automatically construct a database of contents data which are distributed over plural reproducing apparatuses and search this database on the basis of user's fragmentary memory. A contents sharing management system practiced as one embodiment of the invention comprises an episode server installed at user's home and plural reproducing apparatuses including a component stereo set, portable player, portable wireless terminal, and MD player, which are interconnected in a wireless manner based on wireless communication technologies such as Bluetooth. The episode server wirelessly connects to the portable player for example to get the episode information stored therein and organizes the retrieved episode information into a database. The episode server also searches the database upon request from the portable player to identify a source apparatus in which desired contents data are stored and supplies the retrieved contents data to the requesting portable player.

Abstract: A computer-implemented method provides access to multimedia content, which include units of content that include audio components. Meta data for the units of content is formed to an association of key phrases detected in the audio components and the units. In some examples, forming the meta data includes determining a candidate set of key phrases associated with the unit of multimedia and searching for the presence of the candidate key phrases in the audio components. Forming the meta data then includes forming data representing the presence of key phrases in the audio components.

Abstract: A method for searching music based on music segment information inquiry comprises: a) analyzing certain music or song to obtain music rhythm and note information of any segment, and converting it to digital data as a basis for searching the music or the song after quantification; b) storing indexes of any segment of music rhythm and note information for the music or song in database; c) Take the inquiry requirement as a basis for searching and comparing to find the required music or song. The advantage of the invention is to search music via a segment of music melody or song without knowing text information like music name or singer, which extremely extends the flexibility of music searching, and therefore the subscriber's requirements for music searching is satisfied and fuzzy searching is achieved. When searching and comparing, the matching degree between music rhythm and note information in a segment and in index database may be configured to improve searching hit-the-target rate or searching accuracy.

Abstract: This invention description details systems and methods for improving human conversations by enhancing conversation participants' ability to: —Distill out and record core ideas of conversations. —Classify and prioritize these key concepts. —Recollect commitments and issues and take appropriate action. —Analyze and uncover new insight from the linkage of these ideas with those from other conversations.

Abstract: There is disclosed an analyser (101) for building a transcription model (112; 500) using a training database (111) of music. The analyser (101) decomposes the training music (111) into sound events (201a-e) and, in one embodiment, allocates the sound events to leaf nodes (504a-h) of a tree (500). There is also disclosed a transcriber (102) for transcribing music (121) into a transcript (113). The transcript (113) is sequence of symbols that represents the music (121), where each symbol is associated with a sound event in the music (121) being transcribed. In one embodiment, the transcriber (102) associates each of the sound events (201a-e) in the music (121) with a leaf node (504a-h) of a tree (500); in this embodiment the transcript (113) is a list of the leaf nodes (504a-h). The transcript (113) preserves information regarding the sequence of the sound events (201a-e) in the music (121) being transcribed.

Abstract: A computer-implemented method and system for confirming that a remote consultation between a professional and a client occurred, including monitoring a remote consultation call between a professional and a client using a remote consultation system, storing information of the consultation, and confirming that the consultation took place using the stored information. The consultant can be a healthcare professional such as a doctor, and the client can be a patient.

Abstract: A predictive model generator that enhances customer experience, reduces the cost of servicing a customer, and prevents customer attrition by predicting the appropriate interaction channel through analysis of different types of data and filtering of irrelevant data. The model includes a customer interaction data engine for transforming data into a proper format for storage, data warehouse for receiving data from a variety of sources, and a predictive engine for analyzing the data and building models.

Abstract: A system (300), apparatus (200) and method (100) are provided to automatically play/suggest at least one audio accompaniment while a sequence of at least one digital image is being displayed such that the audio accompaniment matches the content of the particular sequence of images and matches any provided and/or generated image metadata. Search terms are derived from the images themselves as well as any metadata provided by the user and these search terms are then used to find audio accompaniment that either (1) contains these search terms or synonyms thereof in the image or associated text (e.g., song text) or (2) represents the sound normally associated with the images, such as rushing water sound for an image of a fast flowing brook. The invention accepts user input, locates appropriate audio accompaniment as search results and presents these results to the user either by playing the audio accompaniment while displaying the images or by suggesting a playlist to the user compiled from these results.