10results about How to "Take up less resources" patented technology

The invention relates to the technical field of voice recognition, in particular to a mixed recognition processing method and device, equipment and a medium, and the method comprises the steps: recognizing command word data based on a command word recognition model, selecting command word data with an error rate lower than a preset range, generating a corresponding phoneme embedding representation, and forming a phoneme embedding dictionary; determining a mixed recognition command word corresponding to the current command word, and generating a mixed recognition list; identifying an audio embedding sequence corresponding to the current command word; obtaining a phoneme embedding sequence of each mixed recognition command word in the mixed recognition list based on a phoneme embedding dictionary; and calculating the similarity between an audio embedding sequence and each phoneme embedding sequence to obtain an identification result. The method and the device can improve recognition accuracy and reduce mixed recognition of command words.

The embodiment of the invention provides a low-complexity signal adaptive detection method and system based on an FPGA, and the method comprises the steps: obtaining a signal envelope which comprises a plurality of data samples; performing amplitude multi-point smoothing processing on the signal envelope to obtain a to-be-judged signal comprising a plurality of judgment samples; obtaining an adaptive threshold value based on the data sample; sequentially comparing the judgment samples with the self-adaptive threshold value; when M continuous judgment samples in the to-be-judged signal are greater than the adaptive threshold values corresponding to the judgment samples, outputting a high-level detection mark; and when N continuous judgment samples in the to-be-judged signal are smaller than the adaptive threshold values corresponding to the judgment samples, outputting a low-level detection mark. Amplitude multi-point smoothing processing is firstly carried out on a signal envelope, then rising edge pulse judgment and falling edge pulse judgment are respectively carried out by adopting multi-data-point accumulation (M-N method), and a relatively high correct judgment probability can also be achieved when the signal-to-noise ratio is relatively low.

The application discloses a method for instruction branch prefetching in an AIGPU processor, which comprises the following steps: 1) defining a branch prefetching instruction in a GPU instruction set, the branch prefetching instruction is used to check whether the instruction at label exists in the cache, if the instruction at label does not exist in the cache, one or more instruction data at label are fetched from the lower storage to the cache; 2) according to the PC absolute value or relative value coded by the branch prefetching instruction, the data at the program branch instruction address is prefetched. The application provides a mechanism for branch prefetching through an instruction, which can reduce the performance loss during branch jumping and improve the system performance. The application does not use a complex branch prediction circuit, and the prediction delay and prediction error are avoided.

The application relates to a method for constructing a data synchronization engine and data query synchronization, a method for constructing a data synchronization engine, which comprises the following steps: step one: configuring three annotations, the annotations comprising: @EnableIaaSDataEngine, @DataInterceptor and @DataBoxHandler; step two: configuring an intercepting processing class interface, which comprises: defining an intercepting processing class interface DataBox, defining processing logic after successful interception, and defining AbstractDataBox; step three: defining an interceptor, parameter management and registering the interceptor; and a method for data query synchronization, which comprises: using the data synchronization engine to query synchronization data. The application solves the problems of data inconsistency and disordered and inharmonious front-end interface display logic caused by switching different underlying IaaS resources under the same management interface.

The present invention discloses a mixing method and device for vehicle audio signals, the mixing method comprising the following steps: A. generating a reference signal according to the vehicle engine speed; B. generating a synthetic signal according to the reference signal, a multimedia signal after filtering and a weight coefficient of noise reduction control; C. generating an error signal according to the synthetic signal and the processed multimedia signal; D. estimating and updating the weight coefficient of noise reduction control according to the error signal; E. generating a mixed signal according to the synthetic signal and the processed multimedia signal. The noise reduction control of the present invention does not perform noise reduction processing on the multimedia signal, thereby ensuring the complete and accurate restoration and playback of the multimedia signal.

The invention relates to the field of satellite navigation and low-orbit satellite navigation enhancement service application, and provides a navigation positioning resolving method based on high, medium and low-orbit mixed constellation fusion. According to the method, the working characteristics and operation modes of low-orbit navigation satellite constellations are fully analyzed according to the requirements of low-orbit navigation satellite receiving and positioning processing engineering implementation, and the dynamic satellite selection method with the optimal geometric configuration is innovatively provided according to the bottlenecks that the low-orbit navigation constellations are large in constellation scale, high in operation speed, large in geometric configuration change and the like. According to the method, positioning participating satellites are selected by taking geometric accuracy factors as a criterion in a geometric accuracy factor pre-evaluation mode, so that the accuracy and robustness of geometric configurations of the positioning participating satellites are ensured, and original satellite parameter support is provided for high, medium and low orbit constellation fusion positioning; the method is of great significance to development of satellite navigation receivers compatible with low-orbit constellations and construction and application of more ubiquitous high, medium and low-orbit collaborative navigation systems in China.

The application discloses a kind of main control board for picosecond laser treatment system, including STM32 single-chip microcomputer and RS232 serial communication touch screen control circuit module, RS232 serial communication laser output control and state monitoring circuit module, opto-coupler isolated IO input circuit module, opto-coupler isolated IO output circuit module, analog voltage input circuit module, TFT card and USB data storage module, DAC circuit module, galvanometer control circuit module and network communication circuit module, the STM32 mainboard is driven by above-mentioned each functional circuit module to the peripheral equipment of corresponding electric control system.The application also discloses 755nm picosecond laser treatment electric control system.The application has big data function and network communication function, has AI auxiliary operation, internet of things system, teaching video function.Greatly improve the treatment effect, safety, reliability, effectiveness of product, reduce product cost, use difficulty, failure rate, maintenance rate and maintenance cost.

The present invention discloses a three-dimensional scene generation method, system, electronic device and storage medium, wherein the three-dimensional scene generation method comprises the steps of: performing lightweight processing on first configuration data to obtain second configuration data; generating a component set based on the second configuration data, wherein the component set includes at least one component; and rendering a target scene based on the component set. The present invention performs lightweight processing on the original configuration data, obtains a component set based on the result of the lightweight processing, and then renders the components in the component set to the target scene, thereby greatly reducing the resources required for three-dimensional scene generation, improving the efficiency of three-dimensional scene generation, integrating different original data into the same target scene, and greatly expanding the relevant application scope of three-dimensional scene generation.

The application provides a digital filtering system for multi-domain analysis of a vector network analyzer, and the specific process of the digital filtering system for processing signals is as follows: multiple analog intermediate frequency input signals enter an analog-digital conversion control and holding unit through a signal conditioning circuit, so as to convert the multiple analog intermediate frequency signals into multiple digital intermediate frequency signals; the multiple digital intermediate frequency signals enter a multi-channel data synchronization unit; under a non-modulation measurement domain, the signals enter a continuous filtering unit, under a synchronous modulation domain, the signals enter a synchronous modulation filtering unit, and when in a data storage mode, a spectrum mode or other special modes, the signals enter an ADC data storage unit; a filtering coefficient generation and control unit generates different FIR filtering coefficients according to input data modes, and sends the filtering coefficients into a multi-channel parallel FIR filtering unit. The technical scheme of the application overcomes the problem that the filtering system in the prior art cannot meet the digital filtering requirements of the vector network analyzer under different measurement modes.

This invention provides a multi-power integrated electrical interlocking device, relating to the field of low-voltage electrical technology. It includes: multiple signal conversion modules and an interlocking connection module. The interlocking connection module includes multiple interlocking connection units corresponding to each signal conversion module. Each interlocking connection unit includes an interlocking wiring network. Each signal conversion module includes one input terminal and multiple output terminals. The input terminal is connected to the status node of a circuit breaker in the multi-power system, and the multiple output terminals are connected to the corresponding interlocking connection unit. Each signal conversion module converts the input signal of the target circuit breaker's status node into multiple sets of dry contact signals. The interlocking wiring network is connected based on the interlocking logic of the corresponding target circuit breaker. The interlocking connection unit performs interlocking control on the corresponding target circuit breaker through the interlocking wiring network and the multiple sets of dry contact signals. This invention achieves forced interlocking between multiple power sources through hardware-level interlocking design without the need for external interlocking lines.