Open Access. Powered by Scholars. Published by Universities.®
- Publication Year
- File Type
Articles 5251 - 5280 of 11865
Full-Text Articles in Entire DC Network
Generative Artificial Intelligence To Provide Seamless Music Transition, Mohsin Alam
Generative Artificial Intelligence To Provide Seamless Music Transition, Mohsin Alam
Defensive Publications Series
Listeners of streamed music may experience brief pauses or interruptions between songs, which can disrupt the listening experience. This disclosure describes techniques for eliminating or reducing pauses between songs on music streaming platforms, thereby providing a seamless, uninterrupted listening experience for users. Artificial intelligence (AI) or machine learning (ML) modules analyze rhythms of the current and upcoming songs. A seamless transition between songs is achieved by determining points in sequential songs that have a common rhythm, or by synthetically generating and inserting a transitional piece of music between sequential songs.
Reducing Wireless Harmonics Using Dynamic Spread Spectrum Clocking And Dynamic Adaptation Of Mipi Frequency, Karl Chiang, Alan Chen, Jehyoung Lee, Nicholas M. Mcdonnell, Andy Hsu, Lendan Lai, Jerry Lai, Chisa Lin
Reducing Wireless Harmonics Using Dynamic Spread Spectrum Clocking And Dynamic Adaptation Of Mipi Frequency, Karl Chiang, Alan Chen, Jehyoung Lee, Nicholas M. Mcdonnell, Andy Hsu, Lendan Lai, Jerry Lai, Chisa Lin
Defensive Publications Series
Harmonics of digital signals generated by components of a mobile device can get coupled to useful signals via the device antenna and appear as noise and desensitize the receiver. This disclosure describes techniques of dynamic spread spectrum clocking (SSC) and dynamic MIPI (mobile industry processor interface) frequency adjustment to arrest the degradation of receiver/antenna performance caused by harmonics generated by device components. Examples of device components include the camera, the MIPI serial interface of the display, the DDIC (display driver integrated circuit) oscillator, etc. Problematic frequency ranges, e.g., frequency ranges of communication channels rich in harmonics generated by components, are …
Augmenting Large Language Models With Audio Generation Capabilities, Ágoston Weisz, Timo Denk, Mauricio Zuluaga, Alessandro Agostini, Christian Frank, Olivier Siegenthaler
Augmenting Large Language Models With Audio Generation Capabilities, Ágoston Weisz, Timo Denk, Mauricio Zuluaga, Alessandro Agostini, Christian Frank, Olivier Siegenthaler
Defensive Publications Series
Chatbots or conversational agent interfaces utilize large language models (LLMs) to provide text responses to user queries. However, such chatbots are not capable of receiving audio input and providing generated audio as a response. This disclosure describes techniques to augment a LLM with an interface to an audio generation model. The LLM is fine-tuned to train it to leverage an API to access the audio generation model when input queries request query response in audio form. The trained LLM performs reasoning tasks and generates prompts for the audio generation model. The user-provided audio input and the LLM-generated prompts are fed …
Smart Payments: Method And System For Integrating Stored Value Wallets With Visa Instalments Service (Vis), Dhruv Mahajan Mr, Habeeb Shaik Khadar Mr, Paridhi Nahar Ms
Smart Payments: Method And System For Integrating Stored Value Wallets With Visa Instalments Service (Vis), Dhruv Mahajan Mr, Habeeb Shaik Khadar Mr, Paridhi Nahar Ms
Defensive Publications Series
The present disclosure relates to a method and a system for integrating stored value wallets with a Visa Instalments Service (VIS). The present disclosure suggests uploading a Bank Identification Number (BIN) file including BINs eligible for an Equated Monthly Installment (EMI) payment and creating specific plans associated with the uploaded BINs. Thereafter, a cardholder selects a digital wallet which includes a stored value wallet for payment during the checkout process on an electronic commerce (e-commerce) merchant website or a mobile application. Once the selection process is successful, the present disclosure suggests initiating a VIS Application Program Interface (API) to facilitate …
Nanoimprint Lithography To Create Variable Slant Angle Waveguides
Nanoimprint Lithography To Create Variable Slant Angle Waveguides
Defensive Publications Series
A headset for virtual reality and augmented reality applications is provided. The headset includes a pixel array of light emitters, and a planar waveguide. The planar waveguide includes a first area on a waveguide including a first grating having multiple grooves formed at a first slant angle, and a second area on the waveguide including a second grating having multiple grooves formed at a second slant angle. The grooves in the first grating are oriented in a first direction, and the grooves in the second grating are oriented in a second direction, and the first grating and the second grating …
Imaging Apparatus For Simultaneously Generating Images At Distinct Wavelengths, N/A
Imaging Apparatus For Simultaneously Generating Images At Distinct Wavelengths, N/A
Defensive Publications Series
Eye-tracking modules in augmented reality devices may use multiple near-infrared illuminators and cameras that flash in an interlaced manner. Due to sequential flashing, the eyeball location captured at a particular time differs between illuminators, thereby reducing tracking accuracy. This disclosure describes techniques for simultaneously generating dual images at distinct near-infrared wavelengths. Illuminators of two near-infrared wavelengths flash simultaneously. An optical apparatus that includes a metalens, a dual-band infrared pass filter, an imaging sensor, etc., separates incoming light into its constituent wavelengths to form simultaneous images at distinct wavelengths. Dual, wavelength-separated images are obtained with little or no difference in captured …
Enforcing Caps On Documents Donated By Apps Via A Donation Api, Keun Soo Yim
Enforcing Caps On Documents Donated By Apps Via A Donation Api, Keun Soo Yim
Defensive Publications Series
Knowledge of user-specific entities donated by various apps can enhance query answering performance of large language models (LLMs). However, system health issues can be caused if a donor app stores an excessive number of entities in the platform storage. This disclosure describes an app-content donation library with built-in mechanisms to prevent or reduce performance, reliability, and quality issues that can arise when third-party apps directly use a low-level donation API. When an app is uploaded to an application store (that makes apps available for installation on the platform), a verification procedure is performed to ensure that the uploaded app archive …
Using Thermal Sensors To Improve Image Realism And To Predict Temperature From Images, Dhiraj Jadhav
Using Thermal Sensors To Improve Image Realism And To Predict Temperature From Images, Dhiraj Jadhav
Defensive Publications Series
An image captured by a camera can have a hue different from that of the actual scene as perceived by the human eye. The difference between the human perception and the image of a scene can be traced to inaccurate capture of the color temperature by the image sensor. This disclosure describes techniques to use temperature sensors on a mobile device or camera to adjust the color temperature of an image to render it realistically. Conversely, an image from a well-calibrated camera can be used to estimate the temperature of the scene. If the camera is capturing live footage, the …
Adaptive Haptic Experience For Existing User Preferences
Adaptive Haptic Experience For Existing User Preferences
Defensive Publications Series
A system for providing an adaptive haptic experience is disclosed. The system may associate a first device with a user. The system may also associate a second device with the user. The system may also determine a first haptic drive waveform associated with the first device. The system may also select, from a plurality of second haptic drive waveforms based on the first haptic drive waveform, a second haptic drive waveform.
An Improved Process For The Preparation Of (N4-(4-([1,2,4]Triazolo[1,5-A]Pyridin-7-Yloxy)-3Methylphenyl)-N6-(4,4-Dimethyl-4,5-Dihydrooxazol-2-Yl)Quinazoline-4,6-Diamine, Msn Laboratories Private Limited, R&D Center; Srinivasan Thirumalai Rajan; Revu Satyanarayana; Nomula Sathaiah; Sadineni Ravi Kumar
An Improved Process For The Preparation Of (N4-(4-([1,2,4]Triazolo[1,5-A]Pyridin-7-Yloxy)-3Methylphenyl)-N6-(4,4-Dimethyl-4,5-Dihydrooxazol-2-Yl)Quinazoline-4,6-Diamine, Msn Laboratories Private Limited, R&D Center; Srinivasan Thirumalai Rajan; Revu Satyanarayana; Nomula Sathaiah; Sadineni Ravi Kumar
Defensive Publications Series
The present invention provides an improved process for the preparation of (N4-(4-([1,2,4]triazolo[1,5-a]pyridin-7-yloxy)-3methylphenyl)-N6-(4,4-dimethyl-4,5-dihydrooxazol-2-yl)quinazoline-4,6-diamine which is referred as Tucatinib
Lossless Compression For Semantic Textures, Joshua Schpok
Lossless Compression For Semantic Textures, Joshua Schpok
Defensive Publications Series
A semantic texture overlays semantic labels over an image to indicate the type of texture represented by each region of the image. Traditional lossy compression works well for color textures, but not for semantic textures. This disclosure describes lossless compression techniques to compress semantic textures, thereby reducing the memory occupied by semantic textures. The techniques leverage the observation that semantic textures, unlike color textures, are highly structured with large blocks of common values. The techniques enable high-speed access to detailed rendering and resolution in real-time computer graphics. They also enable additional textures or texture resolution, enhancing the detail and realism …
Suggesting Actions Based On User Intent Detection Using A Large Language Model, Anant Chowdhary, Ben Johns
Suggesting Actions Based On User Intent Detection Using A Large Language Model, Anant Chowdhary, Ben Johns
Defensive Publications Series
The content displayed on a device screen is often indicative of a user task and, with user permission, can be analyzed to surface suggestions that can enable users to perform such tasks quickly. However, rules-based systems that rely on pre-stored knowledge about text and other content may fail to correctly recognize and surface the appropriate task and need to be modified for different user contexts. This disclosure describes the use of a large language model (LLM), e.g., an on-device model, to analyze on-screen content and its context to automatically determine user intent. The reasoning capabilities of the LLM enable reliable …
An Improved Process For The Preparation Of 4-Amino-5-Chloro-N-[1-(3-Methoxypropyl) Piperidin-4-Yl]-2,3-Dihydrobenzofuran-7-Carboxamide, Msn Laboratories Private Limited, R&D Center , Srinivasan Thirumalai Rajan; Sajja Eswaraiah; Sagyam Rajeshwar Reddy; Sheri Venkata Reddy
An Improved Process For The Preparation Of 4-Amino-5-Chloro-N-[1-(3-Methoxypropyl) Piperidin-4-Yl]-2,3-Dihydrobenzofuran-7-Carboxamide, Msn Laboratories Private Limited, R&D Center , Srinivasan Thirumalai Rajan; Sajja Eswaraiah; Sagyam Rajeshwar Reddy; Sheri Venkata Reddy
Defensive Publications Series
The present invention relates to an improved process for the preparation of 4-amino-5-chloro-N-[1-(3-methoxypropyl)piperidin-4-yl]-2,3-dihydrobenzofuran-7-carboxamide succinate
Event Based Post Transformation Framework For Multi-Task Learning Outputs, Xi Kan, Sheng Wang, Michael Mori, Dan Wang
Event Based Post Transformation Framework For Multi-Task Learning Outputs, Xi Kan, Sheng Wang, Michael Mori, Dan Wang
Defensive Publications Series
Disclosed herein is a method for an event-based post transformation framework for multitask learning outputs in deep neural networks. The method discloses categorization of event types based on specific pre-defined business requirements. Each task associated with multitask learning is then treated as a separate event and grouped or separated according to their correlation with predefined event types. Next, the method calculates event confidence scores based on the probability of raw output scores from each task. The method then determines scaled score cut-offs, which are predetermined based on business needs and in next step utilizes a spline model to transform probability …
Visa Cloud Connect, Brendan Louis, Peter Pozar, Carl Coles, Padmanaba Pillai, Zaheer Babul, Sanjeev Louis Lopes, Roshan Patel, Jonathan Wood, Ogochukwu Ikenyei
Visa Cloud Connect, Brendan Louis, Peter Pozar, Carl Coles, Padmanaba Pillai, Zaheer Babul, Sanjeev Louis Lopes, Roshan Patel, Jonathan Wood, Ogochukwu Ikenyei
Defensive Publications Series
The present disclosure relates to a technique for providing payment processing infrastructure connectivity to cloud-based clients. The technique recites maintaining a virtual datacenter by the client at virtual private cloud hosted by a public cloud provider and a payment processor’s cloud connect is provided by payment processor enabling cloud-based clients to access its network infrastructure. The technique involves transmitting data from client VPC to payment processor’sVPC which in turn further transmits it to payment processor’s point of presence (PoP). The EA servers located at the payment processor’s PoP are accessed by the client via secure private link, eliminating the need …
Automatically Configuring Mobile Device Alerting Mode Based On User Context, Yan Mayster, Bruce Bahnsen
Automatically Configuring Mobile Device Alerting Mode Based On User Context, Yan Mayster, Bruce Bahnsen
Defensive Publications Series
Users need to perform manual operations to modify mobile device alerting settings in many contexts. This disclosure describes techniques that, with user permission, analyze the user’s context such as location, schedule, user’s current level of engagement with the device, etc. to automatically determine an applicable alerting mode for their mobile device. Additionally, data from digital maps regarding location type, location characteristics, etc. as well as real time information regarding events taking place at or near the user’s location can be used in making the determination. Adjusting the alerting mode can include automatically adjusting notification volume levels, timing, and content tailored …
Auto Scaling Of Images With Deep Learning, Hp Inc
Auto Scaling Of Images With Deep Learning, Hp Inc
Defensive Publications Series
Cloud platform supports photo print, scan to email and apps to print/scan. This disclosure provides a solution to detect and do scaling of Images with Deep Learning! Convolutional neural network (CNN) solves the problem of Image scaling in the context of scale-up and scale-down based on resolution. A solution is proposed to automatically detect the image pixels and do the scaling automatically. The algorithm should be able to detect and do scaling with high accuracy while handling documents.
Wi-Fi Mesh-Aware Drone Controller, Robert Barton, Gino Corleto, Matt Swartz, Doosan Jung, Qihong Shao
Wi-Fi Mesh-Aware Drone Controller, Robert Barton, Gino Corleto, Matt Swartz, Doosan Jung, Qihong Shao
Defensive Publications Series
Techniques are presented herein that support a mobile mesh encompassing a constellation of drones. Within such an arrangement, each drone carries a Wi-Fi access point (AP). Aspects of the techniques introduce the concept of a Wi-Fi mesh-aware mobile constellation controller (WFMAMCC) that not only intelligently dispatches Wi-Fi AP carrying drones (to, for example, minimize the number of drones required) but also optimally manages the mesh network that is carried on those drones (to, for example, ensure that, at all times, that mesh network supports a high level of performance and maintains backhaul connectivity). The techniques may find use in any …
Unified Test Suite For Testing Multi-Language Software Development Kits (Sdks), Arpan Mishra, Guang Yang
Unified Test Suite For Testing Multi-Language Software Development Kits (Sdks), Arpan Mishra, Guang Yang
Defensive Publications Series
Cloud computing services provide multilingual software development kits (SDK) to enable customers to interact with the cloud service. Testing the SDK in all of the multiple languages entails an expensive replication of software effort. This disclosure describes a unified test suite, operable in a single language, capable of testing multi-language SDKs. Protocol buffers are used to define a language agnostic interface. Standard tools translate protobufs to SDK languages. Test cases written in a single language interact with the language agnostic interface, rather than directly with the client SDK. A thin adapter layer adapts the protobufs into the corresponding SDK interface, …
Goldpinch Multi Dimensional Scrolling
Goldpinch Multi Dimensional Scrolling
Defensive Publications Series
The present disclosure generally relates to accessing digital content and, more particularly, selecting digital content via a user interface in a social media platform.
Network Topology Reconstruction From Operational Logs, Dmitry Goloubew, Don Allen
Network Topology Reconstruction From Operational Logs, Dmitry Goloubew, Don Allen
Defensive Publications Series
Techniques are presented herein that support a new, low-overhead network topology discovery mechanism that is based on data from existing operational log sources such as, for example, syslog entries. Such an approach lowers the operational cost of any solution that requires knowledge of a network’s topology. Additionally, such an approach allows for the ongoing rediscovery of a network’s topology to track over time the documented or undocumented evolution of the same.
The Light Projection Configuration Which Utilizes An Angularly Oriented Reflective Element To Capture The Light Rays That Are Coming From The Scene In Front Of The User, Where The Captured Light Rays Will Be Projected Onto A Sensor Utilizing A Perpendicularly Oriented Convex Lens Stack, Punarjeewa Abeysekera
Defensive Publications Series
This paper describes a light projection configuration which utilizes an angularly oriented reflective element to capture the light rays that are coming from the scene in front of the user, where the captured light rays will be projected onto a sensor utilizing a perpendicularly oriented convex lens stack. The configuration will contain a flat mirror utilized as the reflective element, 2 convex lenses arranged in the stack configuration and a sensor to capture the light projections coming from the scene. The flat mirror will be configured to be in the angular configuration and will be positioned in front of the …
Portion Of A Display Screen
Defensive Publications Series
Ornamental Design for a PORTION OF A DISPLAY SCREEN
Controlling Smartphone Webcam Mode Using Alternatives To Smartphone Touchscreen, Jayant Chowdhary, Avichal Rakesh, Dipen Patel
Controlling Smartphone Webcam Mode Using Alternatives To Smartphone Touchscreen, Jayant Chowdhary, Avichal Rakesh, Dipen Patel
Defensive Publications Series
This disclosure describes user-friendly techniques to connect a smartphone that is utilized as a web camera (webcam) by pairing the smartphone with a host device. Button-press patterns of a power button or another physical button on the side of a smartphone is used to convey user intent to enable/disable a webcam mode for the smartphone that is connected to a host device. This enables users to turn the webcam mode on/off even when the screen is turned off or facing away from the user, as is the case when a rear camera of the smartphone is used in the webcam …
Semi-Federated Learning Of An Embedding Space Across Multiple Machine Clusters
Semi-Federated Learning Of An Embedding Space Across Multiple Machine Clusters
Defensive Publications Series
Provided are systems and methods for privacy-preserving learning of a shared embedding space for data split across multiple separate clusters of computing machines. In one example, the multiple separate clusters of computing machines can correspond to multiple separate data silos.
Stateful Method For Access Point Discovery Of Wireless Local Area Network Controller, Rahul K Kachalia, Muhammad Imam, Simone Arena
Stateful Method For Access Point Discovery Of Wireless Local Area Network Controller, Rahul K Kachalia, Muhammad Imam, Simone Arena
Defensive Publications Series
Access points (APs) for a wireless local area network (WLAN) can discover a wireless LAN controller (WLC) address (in order to establish a management session with the WLC) through a variety of mechanisms, such as Dynamic Host Configuration Protocol (DHCP) option 43 mechanisms, Domain Name System (DNS) server mechanisms, and Layer 2 (L2) broadcast discovery mechanisms. The DHCP discovery mechanism is the most commonly used mechanism for WLC discovery but is a laborious and manual task that may be prone to errors. Techniques proposed herein provide an easy to use, stateful, and reliable mechanism through which an AP can discover …
Cooktop – Easy Clean Stand With Friction Hinge
Cooktop – Easy Clean Stand With Friction Hinge
Defensive Publications Series
Cooktop is the most common appliance which is used for cooking food on a day to day basis. Cooktop is generally heavy and also occupies a larger area of kitchen table. During cooking, the food stuffs can spill and enter the counter space under the cooktop which needs to be cleaned regularly. The cooktop is generally made of steel or glass. Glass cooktop are generally heavier than steel versions. Lifting the cooktop with one hand and cleaning in the other hand is a pain point. In this paper, an easy lift stand for the cook top which enables the user …
Liquid Metal Antennas For Wearable Devices
Liquid Metal Antennas For Wearable Devices
Defensive Publications Series
The novelty in this invention, rather than being about the liquid metal materials themselves, is around the use of liquid metals as antennas in a stretchable substrate (e.g., the metal could be injected into a silicone band or used as a conductive core of a thread for a textile band). The use of liquid metals as antennas in wearables could be the best way to significantly increase antenna surface area without introducing problematic points of failure, as liquid metals are self-healing (e.g., to minor puncture damage) and highly adaptable. This concept could be used in any wearable strap and may …
Automatically Detecting Expensive Prompts And Configuring Firewall Rules To Mitigate Denial Of Service Attacks On Large Language Models, Assaf Namer, Prashant Kulkarni, Erik Jeansson, Brandon Maltzman, Hauke Vagts
Automatically Detecting Expensive Prompts And Configuring Firewall Rules To Mitigate Denial Of Service Attacks On Large Language Models, Assaf Namer, Prashant Kulkarni, Erik Jeansson, Brandon Maltzman, Hauke Vagts
Defensive Publications Series
Denial of service attacks on generative artificial intelligence systems, e.g., large language models (LLMs), can include sending LLMs requests that include expensive prompts designed to consume computing resources and degrade model performance. This disclosure describes techniques to automatically detect such prompts and then configure firewall rules that prevent such prompts in subsequent requests from reaching the LLM. Per the techniques, prompts provided to an LLM are matched against input and output token size as well as resource utilization to identify prompts that deviate significantly from a baseline. Expensive prompts are identified, and semantically similar prompts are automatically generated using the …
Customer Feedback Segmentation, Summarization, And Natural Language Querying Using Machine Learning, Yuanzhi Xu
Customer Feedback Segmentation, Summarization, And Natural Language Querying Using Machine Learning, Yuanzhi Xu
Defensive Publications Series
It is important for businesses to have an insightful understanding of their customers' needs. It is often difficult to extract statistically meaningful and actionable insights from customer feedback. This disclosure describes the use of machine learning techniques to automatically summarize customer feedback and to generate insights from such summaries using natural language queries. The techniques scale with business size and product volume such that top of mind product issues can be identified from unstructured feedback. By enabling scalable examination, categorization, and natural language querying of customer feedback, the described techniques can help a business discover customer pain points. Raw customer …