Abstract: Utilization data of reserved virtual machine (VM) instances can be obtained from a cloud service provider. A refund recommendation comprising a first quantity of VM instances and a refund credit associated with the first quantity of VM instances can be generated based on the utilization data. A purchase recommendation comprising a second quantity of VM instances and a purchase cost associated with the second quantity of VM instances can be generated. An exchange transaction to return the first quantity of VM instances and purchase the second quantity of VM instances can be generated, where a net total of the exchange transaction is determined based on the refund credit and the purchase cost.

Abstract: Technologies for safely lowering a DC link voltage potential include detecting shut down of a system that is powered by the DC link energy storage system and initiating an operating mode to dissipate energy as a form of loss without utilizing an additional resistor, that is, dissipating the DC link internally to the enclosed power module.

Abstract: The present disclosure provides a test system and a method for measuring efficiency of underground coal gasification, and relates to the technical field of underground coal gasification, comprising a gasification agent supply system, a gasifier, an outlet and recycling system and a monitoring system, wherein the gasification agent is mainly injected into the gasifier and then chemically reacts with the coal by igniting the gasifier at a high temperature to generate gas, while after the gas passes through the outlet and recycling system, the composition and content are measured, and the monitoring system monitors an operating state of the gasifier, injection of the gasification agent and production of gas, temperature change in the gasification channel and ignition temperature and pressure gauge data.

Abstract: A battery module, a battery pack, and a vehicle are provided. The battery module includes a battery body, an end plate, and a cover body. The battery body includes multiple battery cells. The multiple battery cells are arranged side by side. The end plate is arranged on an end portion of the battery body. According to the battery module, the end plate is arranged on the end portion of the battery body, the end plate is provided with a first exhaust channel therein, the cover body is engaged with multiple pressure relief valves to form a second exhaust channel, and the second exhaust channel is in communication with the first exhaust channel.

Abstract: Technologies for safely lowering a DC link voltage potential include detecting shut down of a system that is powered by the DC link energy storage system and initiating an operating mode to dissipate energy as a form of loss without utilizing an additional resistor, that is, dissipating the DC link internally to the enclosed power module.

Abstract: An application management method includes collecting a plurality of objects in an application example used for serverless computing and generating a base tree according to a reference hierarchical relationship of the plurality of objects; performing internal memory object rearrangement on a plurality of application examples according to the base tree; and performing internal memory merging on the plurality of rearranged application examples.

Abstract: An automatic speech recognition system and a method thereof are provided. The system includes an encoder and a decoder. The encoder comprises a plurality of encoder layers. At least one encoder layer includes a plurality of encoder sublayers fused into one or more encoder kernels. The system further comprises a first pair of ping-pong buffers communicating with the one or more encoder kernels. The decoder comprises a plurality of decoder layers. At least one decoder layer includes a plurality of decoder sublayers fused into one or more decoder kernels. The decoder receives a decoder output related to the encoder output and generates a decoder output. The encoder sends the decoder output to a beam search kernel.

Abstract: A method of reducing host nucleic acids in a biological sample and a use thereof, which belong to the technical field of gene detection is provided. The method includes the steps of a) performing pre-treatment for the biological sample under a mild condition with partial lysis and/or no lysis of host cell membrane, thereby obtaining a liquid sample; and b) taking the liquid sample followed by adding a nucleic acid digestion reagent, thereby degrading the host nucleic acid which is exposed in the liquid sample. The method is used for reducing host nucleic acids in a biological sample and non-selectively enriching the targets, such as bacteria, fungi, viruses, mycoplasma/chlamydia.

Abstract: Technologies for safely lowering a DC link voltage potential include detecting shut down of a system that is powered by the DC link energy storage system and initiating an operating mode to dissipate energy as a form of loss without utilizing an additional resistor, that is, dissipating the DC link internally to the enclosed power module.

Abstract: The present disclosure relates to an automatic speech recognition system and a method thereof. The system includes a conformer encoder and a pair of ping-pong buffers. The encoder includes a plurality of encoder layers sequentially executed by one or more graphic processing units. At least one encoder layer includes a first feed forward module, a multi-head self-attention module, a convolution module, and a second feed forward module. The convolution module and the multi-head self-attention module are sandwiched between the first feedforward module and the second feed forward module. The four modules respectively include a plurality of encoder sublayers fused into one or more encoder kernels. The one or more encoder kernels respectively read from one of the pair of ping-pong buffers and write into the other of the pair of ping-pong buffers.

Abstract: Disclosed are a method, device and readable storage medium for speech recognition. The method includes: determining speech features of the speech data by feature extraction on the speech data; determining syllable data corresponding to each of the speech features based on a plurality of feature extraction layers and a softmax function layer included in an acoustic model, where the acoustic model is configured to convert the speech feature into the syllable data; determining text data corresponding to the speech data based on a language model, a pronouncing dictionary and the syllable data, where the pronouncing dictionary is configured to convert the syllable data into the text data, and the language model is configured to evaluate the text data; and outputting the text data.

Abstract: Provided are a liquid crystal alignment agent, a liquid crystal alignment film and a liquid crystal display element. The liquid crystal alignment agent at least comprises one of a polyamic acid solution and a polyimide solution. The polyamic acid solution is obtained by means of a polymerization reaction of diamine compound component A and dianhydride compound component B in a solvent, and the polyimide solution is obtained from the polyamic acid solution via a dehydration imidization treatment. The diamine component A at least comprises one of the diamine compounds represented by “formula 1”. The obtained liquid crystal alignment agent has a better liquid crystal alignment capability, an excellent effect on image sticking, and the effect that decomposition products resulting from light alignment are easy to remove.

Abstract: Methods and systems for controlling a multipurpose power converter for converting power for a transport climate control system are provided. The multipurpose power converter includes a rectifier having a first leg, a second leg, and a third leg. The multipurpose power converter also includes a first switch, a second switch, and an inductor-capacitor network. The first switch and the second switch are connected to the third leg. The inductor-capacitor network is connected to the first switch. When the first switch is on and the second switch is off, the multipurpose power converter is configured as a single-phase AC power converter. When the first switch is off and the second switch is on, the multipurpose power converter is configured as a three-phase AC power converter.

Abstract: An automatic speech recognition system and a method thereof are provided. The system includes an encoder and a decoder. The encoder comprises a plurality of encoder layers. At least one encoder layer includes a plurality of encoder sublayers fused into one or more encoder kernels. The system further comprises a first pair of ping-pong buffers communicating with the one or more encoder kernels. The decoder comprises a plurality of decoder layers. At least one decoder layer includes a plurality of decoder sublayers fused into one or more decoder kernels. The decoder receives a decoder output related to the encoder output and generates a decoder output. The encoder sends the decoder output to a beam search kernel.

Abstract: The present disclosure relates to an automatic speech recognition system and a method thereof. The system includes a conformer encoder and a pair of ping-pong buffers. The encoder includes a plurality of encoder layers sequentially executed by one or more graphic processing units. At least one encoder layer includes a first feed forward module, a multi-head self-attention module, a convolution module, and a second feed forward module. The convolution module and the multi-head self-attention module are sandwiched between the first feedforward module and the second feed forward module. The four modules respectively include a plurality of encoder sublayers fused into one or more encoder kernels. The one or more encoder kernels respectively read from one of the pair of ping-pong buffers and write into the other of the pair of ping-pong buffers.

Abstract: Provided are a liquid crystal alignment agent, a liquid crystal alignment film and a liquid crystal display element. The liquid crystal alignment agent at least comprises one of a polyamic acid solution and a polyimide solution. The polyamic acid solution is obtained by means of a polymerization reaction of diamine compound component A and dianhydride compound component B in a solvent, and the polyimide solution is obtained from the polyamic acid solution via a dehydration imidization treatment. The diamine component A at least comprises one of the diamine compounds represented by “formula 1”. The obtained liquid crystal alignment agent has a better liquid crystal alignment capability, an excellent effect on image sticking, and the effect that decomposition products resulting from light alignment are easy to remove.

Abstract: Methods and systems for controlling a multipurpose power converter for converting power for a transport climate control system are provided. The multipurpose power converter includes a rectifier having a first leg, a second leg, and a third leg. The multipurpose power converter also includes a first switch, a second switch, and an inductor-capacitor network. The first switch and the second switch are connected to the third leg. The inductor-capacitor network is connected to the first switch. When the first switch is on and the second switch is off, the multipurpose power converter is configured as a single-phase AC power converter. When the first switch is off and the second switch is on, the multipurpose power converter is configured as a three-phase AC power converter.

Abstract: Disclosed are a method, an electronic device, and a computer readable storage medium for creating a vote. The method includes: receiving a vote initiation instruction to obtain voice information inputted currently; determining textual information by recognizing the voice information; determining at least two option words based on semantic recognition of the textual information; generating a question stem of the vote based on the textual information; and generating voting options based on the option words for the vote.

Abstract: A method and device for generating speech recognition model are provided. The method includes: obtaining training samples, wherein each training sample includes a speech frame sequence and a labeled text sequence; training the encoder by using the speech frame sequence as an input feature and using speech encoded frames of the speech frame sequence as an output feature; training the decoder by using the speech encoded frames as a first input feature and using the labeled text sequence as a first output feature, and obtaining a current prediction text sequence; and training the decoder again by using the speech encoded frames as a second input feature and using a sequence as a second output feature, wherein the sequence is obtained by sampling the labeled text sequence and the current prediction text sequence based on a preset probability.

Abstract: Disclosed are a method, device and readable storage medium for speech recognition. The method includes: determining speech features of the speech data by feature extraction on the speech data; determining syllable data corresponding to each of the speech features based on a plurality of feature extraction layers and a softmax function layer included in an acoustic model, where the acoustic model is configured to convert the speech feature into the syllable data; determining text data corresponding to the speech data based on a language model, a pronouncing dictionary and the syllable data, where the pronouncing dictionary is configured to convert the syllable data into the text data, and the language model is configured to evaluate the text data; and outputting the text data.