Abstract: A computer system can unify main memory and cache memory, wherein fully associative mapping method can be utilized to cover a whole range of cache and main memory. In the system, central processing unit (CPU) sends a data request and access the cache portion of the unified cache and memory system; fully associative search is conducted on the unified cache and main memory system as one range of physical memory; if matching data is found on the cache portion, the data is returned to the CPU; if matching data is found on the main memory portion, the matching data is swapped to the cache portion and then return the data to the CPU; if matching data is not found in either portion of the unified cache and main memory system, the operating system is trigged to handle the page fault.

Abstract: A computer-implemented intent classification method and system are provided. The method comprises: providing a plurality of trained model sets, each based on a different model and comprising a plurality of trained models, with each trained model in each trained model set obtained by training a same model with a different initialization condition over a pre-determined dataset; feeding an input text into each trained model to obtain a plurality of prediction results for indicating a likely intent class of the input text; ensembling prediction results corresponding to a same trained model set for each trained model set to obtain a plurality of first-layer ensembles; ensembling the plurality of first-layer ensembles to obtain at least one second-layer ensemble, each obtained by ensembling two or more first-layer ensembles; and determining the intent class of the input text based on the at least one second-layer ensemble.

Abstract: An intent classification method and a system for realizing the intent classification method are provided. The intent classification method can be used for determining an intent class of a sentence input in a language containing information-bearing characters, such as Chinese. The method specifically comprises a step of obtaining at least one set of word embeddings and character embeddings from the sentence input to thereby generate at least one set of integrated word-character embeddings, and a step of determining an intent class corresponding to the sentence input based on the at least one set of integrated word-character embeddings. Herein, each of the at least one set of integrated word-character embeddings corresponds to one of the at least one set of word embeddings and character embeddings.

Abstract: A human augmented robotics intelligence operation system can include a plurality of robots, each robot having a plurality of sensors; a robot control unit; and one or more articulating joints; a cloud-based robotic intelligence engine having; a communication module; a historical database; and a processor; and a human augmentation platform. The processor can be configured to make a probabilistic determination regarding the likelihood of successfully completing the particular user command. When the probabilistic determination is above a pre-determined threshold, the processor sends necessary executable commands to the robot control unit. Alternatively, when the probabilistic determination is below the predetermined threshold, the processor generates an alert and flags the operation for human review.

Abstract: An audio-visual perception system includes an audio perception module and a visual perception module, respectively receiving sounds and images of a subject and converting them into audio and visual signals. At least one of the above two modules is a first perception module having more than one working mode. A processing and control module controls the first perception module to switch a working mode thereof based on the audio or visual signals received from one of the two above modules other than the first perception module. An audio-visual perception apparatus having the audio-visual perception system is also disclosed, which can be used as a robot control unit (RCU) mounted onto a robot, allowing a human-in-the-loop robot operator to visually and audibly monitor the subject in a surrounding of the robot. The RCU can switch between a RCU mode and a cell phone mode.

Abstract: A speech recognition method includes: providing a system having a local computational device, the local computational device having a microphone, processing circuitry, and a non-transitory computer-readable medium; recording a raw audio waveform utilizing the microphone; determining a background noise condition for the raw audio waveform; comparing the background noise condition to a plurality of linguistic models having associated background noise conditions; determining a nearest match between the background noise condition of the raw audio waveform and the associated background noise condition of at least one of the plurality of linguistic models; and performing an automatic speech recognition (ASR) function between the raw audio waveform and the linguistic model having the matching associated background noise condition.

Abstract: A voice activity detection method includes: training one or more computerized neural networks having a denoising autoencoder and a classifier, wherein the training is performed utilizing one or more models including Mel-frequency cepstral coefficients (MFCC) features, ? features, ?? features, and Pitch features, each model being recorded at one or more differing associated predetermined signal to noise ratios; recording a raw audio waveform and transmitting the raw audio waveform to the computerized neural network; denoising the raw audio wave utilizing the denoising autoencoder; and determining whether the raw audio waveform contains human speech; extracting any human speech from the raw audio waveform.

Abstract: A computer-implemented intent classification method and system are provided. The method comprises: providing a plurality of trained model sets, each based on a different model and comprising a plurality of trained models, with each trained model in each trained model set obtained by training a same model with a different initialization condition over a pre-determined dataset; feeding an input text into each trained model to obtain a plurality of prediction results for indicating a likely intent class of the input text; ensembling prediction results corresponding to a same trained model set for each trained model set to obtain a plurality of first-layer ensembles; ensembling the plurality of first-layer ensembles to obtain at least one second-layer ensemble, each obtained by ensembling two or more first-layer ensembles; and determining the intent class of the input text based on the at least one second-layer ensemble.

Abstract: An intent classification method and a system for realizing the intent classification method are provided. The intent classification method can be used for determining an intent class of a sentence input in a language containing information-bearing characters, such as Chinese. The method specifically comprises a step of obtaining at least one set of word embeddings and character embeddings from the sentence input to thereby generate at least one set of integrated word-character embeddings, and a step of determining an intent class corresponding to the sentence input based on the at least one set of integrated word-character embeddings. Herein, each of the at least one set of integrated word-character embeddings corresponds to one of the at least one set of word embeddings and character embeddings.

Abstract: A natural language understanding (NLU) system used in a dialogue systems comprises a first-level NLU sub-system and at least one second-level NLU sub-system. Each second-level NLU sub-system is communicatively coupled with, and has a relatively higher performance than, the first-level NLU sub-system. The first-level NLU sub-system performs a first calculation over an input text received, and then outputs a first meaning if the first meaning is generated with a first confidence level surpassing a first threshold or passes on the input text to one second-level NLU sub-system based on a pre-determined rule if otherwise. Each second-level NLU sub-system receives the input text from the first-level NLU sub-system, and performs a second calculation over the input text, and then outputs a second meaning if the second meaning is generated with a second confidence level surpassing a second threshold or outputs a result indicating a rejection of meaning if otherwise.

Abstract: A data annotation apparatus for machine learning is provided, which includes a stimulus generation portion, a biometrics reading portion, and a data integration portion. The stimulus generation portion is configured to generate, and present to an agent, at least one stimulus based on a first data from a first machine learning dataset. The biometrics reading portion is configured to measure at least one response of the agent to the at least one stimulus, and to generate biometrics data based on the at least one response. The data integration portion is configured to integrate the biometrics data, data of the at least one stimulus, and data of the first machine learning dataset to thereby obtain a second machine learning dataset. The data annotation apparatus can result in an improved data labeling and an enhanced machine learning.