Abstract: There is provided a speech synthesis system and method with an adjustable utterance length. The speech synthesis method according to an embodiment predicts a duration of each phoneme corresponding to a speech mask from the speech mask and a text to be synthesized with the speech mask, encodes the text to be synthesized and extracts a text sequence which is expressed by feature information of the text, generates a speech frame sequence by regulating a length of each phoneme of the text sequence according to the predicted duration of each phoneme corresponding to the speech mask, and synthesizes a speech from the generated speech frame sequence. Accordingly, a length of a speech to be synthesized can be freely regulated as a user desires by regulating a length of a speech mask.

Abstract: There is provided a training dataset construction method for speech synthesis through fusion of language, speaker, emotion within an utterance. A training dataset construction method of a speech synthesis model according to an embodiment collects speech data having different speech utterance information, increases the speech data by fusing the collected speech data within one utterance, and generates a training dataset by using the increased speech data. Accordingly, a training dataset for speech synthesis is constructed through fusion of language, speaker, emotion within one utterance, so that quality of speech synthesis of multi-speaker/multi-language/emotion can be enhanced.

Abstract: There is provided a training method of a multi-task integrated deep learning model. A multi-task integrated deep learning model training method according to an embodiment may generate training data for a plurality of visual intelligence tasks from visual data in a batch, and may train a multi-task integrated deep learning model which performs a plurality of visual intelligence tasks by using the generated training data. Accordingly, training data for training an integrated deep learning model which performs various visual intelligence tasks is generated in a batch through multi-data conversion kernels, so that appropriate training data for performing multiple tasks may be easily obtained and effective training of a multi-task integrated deep learning model is possible.

Abstract: The present invention relates to an apparatus and method capable of estimating a battery state of health, and more particularly, provides an apparatus and method for more accurately calculating the state of balance by voltage of a battery to reflect SOC information as well as the voltage distribution of the battery, and measuring the state of health of the battery on the basis of the state of balance by voltage.

Abstract: The present invention relates to a waste battery discharge process for recycling a waste battery, and more particularly, to a process for checking stability, a state of charge, and a remaining life of the waste battery before reusing or recycling the waste battery (battery after use), and safely and cost-effectively discharging the waste battery.

Abstract: An electronic device may have a display with touch sensors. One or more shielding layers may be interposed between the display and the touch sensors. The shielding layers may include shielding structures such as a conductive mesh structure and/or a transparent conductive film. The shielding structures may be actively driven or passively biased. In the active driving scheme, one or more inverting circuits may receive a noise signal from a cathode layer in the display and/or from the shielding structures, invert the received noise signal, and drive the inverted noise signal back onto the shielding structures to prevent any noise from the display from negatively impacting the performance of the touch sensors. In the passive biasing scheme, the shielding structures may be biased to a power supply voltage.

Abstract: The present disclosure relates to a MoS2-Ni electrode and, in more detail, to a method for manufacturing an alkaline electrolysis electrode through co-sputtering, improves a 1T structure yield through a single process, and provides an electrolysis electrode with high activity in alkaline water electrolysis.

Abstract: A display panel includes an amorphous silicon gate driver in which a lower voltage than the gate-off voltage output from the gate driver is applied to an adjacent stage as a low voltage transmission signal.

Abstract: The present disclosure discloses a computing-aware traffic steering method, apparatus and system in a mobile network. According to the present disclosure, there is provided a computing-aware traffic steering method of mobile network control plane, the method including: in a case where it is confirmed that there is no target data network access identifier (DNAI) mapped to a service ID when a PDU session establishment request including the service ID of a user terminal (UE) is made, transmitting the service ID and user plane path information to a computing-aware traffic steering mobile user plane controller (CATS-MUP-C) using segment routing; reconfiguring a user plane path for accessing an Anycast IP address corresponding to the service ID in conjunction with the CATS-MUP-C; and completing the PDU session establishment using the Anycast IP address.

Abstract: An ultra-micro electrode that performs electrochemical analysis and a method of manufacturing the same is provided. The method includes preparing an insulating member having a capillary shape with a hollow region, preparing a conductive wire including a conductive material with an oxidation resistance greater than an oxidation resistance of copper, inserting the conductive wire into the hollow region of the insulating member, manufacturing half-finished products by applying heat and an external force to the insulating member having the conductive wire inserted therein to stretch the insulating member and the conductive wire so that a central portion of the insulating member is thin and the insulating member and the conductive wire are ruptured at central portions thereof, forming a groove by etching the conductive wire located on a ruptured surface of each half-finished product, and forming a filling layer by filling the groove with a conductive material by a plating process.

Abstract: An electrode manufacturing system includes a transferring part, a coating part, a drying part, and a heat treatment part. The transferring part is arranged downstream of the drying part to heat the electrode substrate to a high temperature and includes a transfer roller supports and transfers an electrode substrate in a long-sheet shape. The coating part applies an electrode slurry to an electrode current collector. The drying part dries the electrode slurry disposed on the electrode substrate. The heat treatment part is disposed downstream of the drying part along a transfer direction of the electrode substrate and applies heat to the electrode substrate to increase the adhesion of the electrode by changing the crystal phase of the PVdF binder of the electrode substrate during the heat treatment process, thereby preventing the separation of the electrode active material layer. A manufacturing method including the same is also provided.

Abstract: The present disclosure discloses a computing-aware traffic steering method, apparatus and system in a mobile network. According to the present disclosure, there is provided a computing-aware traffic steering method of mobile network control plane, the method including in a case where it is confirmed that there is no target data network access identifier (DNAI) mapped to a service ID when a PDU session establishment request including the service ID of a user equipment (UE) is made, transmitting the service ID and user plane path information to a CATS controller in a computing-aware traffic steering (CATS) network environment using segment routing, and reconfiguring a user plane path for accessing an optimal service contact instance IP address determined as one of a plurality of service contact instances based on a base station connected by the UE and computing and networking metrics in conjunction with the CATS controller.

Abstract: A service mesh traffic management apparatus, system, and method in a distributed Kubernetes multi-cluster environment are disclosed. A service mesh traffic management apparatus in a distributed Kubernetes multi-cluster environment comprises a processor, and a memory connected to the processor, wherein the memory stores program instructions executable by the processor for performing operations comprising receiving a monitoring metric collected by a Prometheus server distributed in the multi-cluster, selecting a first routing policy based on a predefined routing policy selection rule and the collected monitoring metric, comparing the first routing policy and a second routing policy currently in use, distributing, when the first routing policy and the second routing policy are different, the first routing policy to each proxy in the service mesh.

Abstract: Provided is a method for managing an address for an application of a LISP (Locator ID Separation Protocol) network system in a distributed container-based virtualization environment, and the method comprises detecting, by a service discovery agent of a container platform, generation of an application in a corresponding cluster; querying, by the service discovery agent, a service IP address to be assigned to an application to be generated by transmitting a service name to a central control plane; searching, by the control plane, for a service IP address matching the service name and transmitting the service IP address to the service discovery agent; and transmitting, by the service discovery agent, the service IP address to a manager of the container platform to complete the generation of the application, and mapping the service IP address to a public IP for the generated application to register it in the control plane.

Abstract: A map generation method including receiving a map editing request for a specific floor among a plurality of floors of a building, providing an editing interface on a display unit of an electronic device in response to the map editing request, the editing interface including at least a part of a specific map corresponding to the specific floor, specifying at least one node group allocatable on the specific map based on first node rules, the first node rules corresponding to spatial characteristics of the specific floor, and performing a node placement process such that first nodes included among the at least one node group are placed on the specific map.

Abstract: A map generation method and system for robot operation. The map generation method including receiving a map editing request for a specific floor among a plurality of floors of a building, providing an editing interface on a display unit of an electronic device in response to the map editing request, the editing interface including at least a part of a specific map corresponding to the specific floor, allocating at least one graphic object on the specific map based on editing information received from the electronic device, and updating the specific map on a cloud server based on completion of the allocating such that robots travel through the specific floor according to an attribute of the at least one graphic object.

Abstract: The present invention provides a system and method for measuring the load of a vehicle using an optimal resonant frequency search wireless power transmission system, wherein a wireless power transmission system, which has maximum power transmission efficiency due to effectively searching for an optimal resonant frequency by varying system frequency, is provided in order to supply wireless power to various sensors mounted on wheels and wheel housings, etc. of the vehicle, and the load can be accurately and effectively measured by using the wireless power transmission system without additional sensors or devices.

Abstract: Disclosed is a network-on-chip including a first data converter that receives first image data and second image data from at least one image sensor and encodes one image data among the first image data and the second image data, into first data, based on whether the first image data is identical to the second image data and a second data converter that receives non-image data from at least one non-image sensor and encodes the received non-image data into second data. The network-on-chip outputs the first data and the second data to transmit the first data and the second data to an external server at a burst length.

Abstract: The present invention provides a system and method for measuring the load of a vehicle using an optimal resonant frequency search wireless power transmission system, wherein a wireless power transmission system, which has maximum power transmission efficiency due to effectively searching for an optimal resonant frequency by varying system frequency, is provided in order to supply wireless power to various sensors mounted on wheels and wheel housings, etc. of the vehicle, and the load can be accurately and effectively measured by using the wireless power transmission system without additional sensors or devices.

Abstract: A method for manufacturing a lightweight cowl crossbar includes: producing an inner pipe, laminating a plurality of composite material layers wound around the inner pipe, and extruding an outer pipe on the winding layer, in which the plurality of composite material layers adjacent to each other are wound in different directions.