Abstract: Systems and methods are provided for obtaining, training, and using an end-to-end AST model based on a neural transducer, the end-to-end AST model comprising at least (i) an acoustic encoder which is configured to receive and encode audio data, (ii) a prediction network which is integrated in a parallel model architecture with the acoustic encoder in the end-to-end AST model, and (iii) a joint layer which is integrated in series with the acoustic encoder and prediction network. The end-to-end AST model is configured to generate a transcription in the second language of input audio data in the first language such that the acoustic encoder learns a plurality of temporal processing paths.

Abstract: A computer implemented method includes receiving speech data representative of speech in a first language The speech data is divided into chunks of speech data, each chunk comprising multiple temporally consecutive frames of acoustic information. Each temporally consecutive chunk of data is processed using beam search on each frame to identify candidate language tokens representing a second language different from the first language. A best candidate language token(s) is selected for each chunk as processed. The selected best candidate language token or tokens for each chunk of data is committed as a prefix for a next temporally consecutive chunk of data.

Abstract: Systems and methods are provided for obtaining, training, and using an end-to-end AST model based on a neural transducer, the end-to-end AST model comprising at least (i) an acoustic encoder which is configured to receive and encode audio data, (ii) a prediction network which is integrated in a parallel model architecture with the acoustic encoder in the end-to-end AST model, and (iii) a joint layer which is integrated in series with the acoustic encoder and prediction network. The end-to-end AST model is configured to generate a transcription in the second language of input audio data in the first language such that the acoustic encoder learns a plurality of temporal processing paths.

Abstract: A method of attention-based end-to-end (A-E2E) automatic speech recognition (ASR) training, includes performing cross-entropy training of a model, based on one or more input features of a speech signal, determining a posterior probability vector at a time of a first wrong token among one or more output tokens of the model of which the cross-entropy training is performed, and determining a loss of the first wrong token at the time, based on the determined posterior probability vector. The method further includes determining a total loss of a training set of the model of which the cross-entropy training is performed, based on the determined loss of the first wrong token, and updating the model of which the cross-entropy training is performed, based on the determined total loss of the training set.

Abstract: A method of attention-based end-to-end (A-E2E) automatic speech recognition (ASR) training, includes performing cross-entropy training of a model, based on one or more input features of a speech signal, determining a posterior probability vector at a time of a first wrong token among one or more output tokens of the model of which the cross-entropy training is performed, and determining a loss of the first wrong token at the time, based on the determined posterior probability vector. The method further includes determining a total loss of a training set of the model of which the cross-entropy training is performed, based on the determined loss of the first wrong token, and updating the model of which the cross-entropy training is performed, based on the determined total loss of the training set.

Abstract: A method of attention-based end-to-end (A-E2E) automatic speech recognition (ASR) training, includes performing cross-entropy training of a model, based on one or more input features of a speech signal, determining a posterior probability vector at a time of a first wrong token among one or more output tokens of the model of which the cross-entropy training is performed, and determining a loss of the first wrong token at the time, based on the determined posterior probability vector. The method further includes determining a total loss of a training set of the model of which the cross-entropy training is performed, based on the determined loss of the first wrong token, and updating the model of which the cross-entropy training is performed, based on the determined total loss of the training set.

Abstract: A method for selecting an optimal language model weight (LMW) used to perform automatic speech recognition, including decoding test audio into a lattice using a language model; analyzing the lattice using a first LMW of a plurality of LMWs to determine a first plurality of best paths; analyzing the lattice using a second LMW of the plurality of LMWs to determine a second plurality of best paths; determining a first best path change rate (BCPR) corresponding to the first LMW based on a number of best path changes between the first plurality of best paths and the second plurality of best paths; and determining the first LMW to be the optimal LMW based on the first BCPR being a lowest BCPR from among a plurality of BCPRs corresponding to the plurality of LMWs.

Abstract: A method of attention-based end-to-end (E2E) automatic speech recognition (ASR) training, includes performing cross-entropy training of a model, based on one or more input features of a speech signal, performing beam searching of the model of which the cross-entropy training is performed, to generate an n-best hypotheses list of output hypotheses, and determining a one-best hypothesis among the generated n-best hypotheses list. The method further includes determining a character-based gradient and a word-based gradient, based on the model of which the cross-entropy training is performed and a loss function in which a distance between a reference sequence and the determined one-best hypothesis is maximized, and performing backpropagation of the determined character-based gradient and the determined word-based gradient to the model, to update the model.

Abstract: An embodiment of a laser induced breakdown system is described that comprises a portable device that includes: a laser configured to produce a beam comprising a plurality of repeating pulses; a processor configured to open a data acquisition window after a delay period, wherein the delay period begins upon production of one of the pulses; one or more optical elements configured to direct the beam at a sample and collect emitted light from a plasma continuum; and an optical detector configured to produce a plurality of signal values from the emitted light from the plasma continuum collected during the data acquisition window, wherein the processor is configured to identify an element from the signal values.

Abstract: A method of attention-based end-to-end (A-E2E) automatic speech recognition (ASR) training, includes performing cross-entropy training of a model, based on one or more input features of a speech signal, determining a posterior probability vector at a time of a first wrong token among one or more output tokens of the model of which the cross-entropy training is performed, and determining a loss of the first wrong token at the time, based on the determined posterior probability vector. The method further includes determining a total loss of a training set of the model of which the cross-entropy training is performed, based on the determined loss of the first wrong token, and updating the model of which the cross-entropy training is performed, based on the determined total loss of the training set.

Abstract: A method for selecting an optimal language model weight (LMW) used to perform automatic speech recognition, including decoding test audio into a lattice using a language model; analyzing the lattice using a first LMW of a plurality of LMWs to determine a first plurality of best paths; analyzing the lattice using a second LMW of the plurality of LMWs to determine a second plurality of best paths; determining a first best path change rate (BCPR) corresponding to the first LMW based on a number of best path changes between the first plurality of best paths and the second plurality of best paths; and determining the first LMW to be the optimal LMW based on the first BCPR being a lowest BCPR from among a plurality of BCPRs corresponding to the plurality of LMWs.

Abstract: A method of attention-based end-to-end (E2E) automatic speech recognition (ASR) training, includes performing cross-entropy training of a model, based on one or more input features of a speech signal, performing beam searching of the model of which the cross-entropy training is performed, to generate an n-best hypotheses list of output hypotheses, and determining a one-best hypothesis among the generated n-best hypotheses list. The method further includes determining a character-based gradient and a word-based gradient, based on the model of which the cross-entropy training is performed and a loss function in which a distance between a reference sequence and the determined one-best hypothesis is maximized, and performing backpropagation of the determined character-based gradient and the determined word-based gradient to the model, to update the model.

Abstract: An embodiment of a laser induced breakdown system is described that comprises a portable device that includes: a laser configured to produce a beam comprising a plurality of repeating pulses; a processor configured to open a data acquisition window after a delay period, wherein the delay period begins upon production of one of the pulses; one or more optical elements configured to direct the beam at a sample and collect emitted light from a plasma continuum; and an optical detector configured to produce a plurality of signal values from the emitted light from the plasma continuum collected during the data acquisition window, wherein the processor is configured to identify an element from the signal values.

Abstract: A method for configuring level shifter spare cells includes providing a power rail connectable to a corresponding power domain, and providing a spare cell including a level shifter circuit. The level shifter circuit has first and second terminals that are connectable to the power rail, and the first and second terminals are floating with respect to the power rail.

Abstract: An embodiment of a laser induced breakdown system is described that comprises a portable device that includes: a laser configured to produce a beam comprising a plurality of repeating pulses; a processor configured to open a data acquisition window after a delay period, wherein the delay period begins upon production of one of the pulses; one or more optical elements configured to direct the beam at a sample and collect emitted light from a plasma continuum; and an optical detector configured to produce a plurality of signal values from the emitted light from the plasma continuum collected during the data acquisition window, wherein the processor is configured to identify an element from the signal values.

Abstract: An isolation cell clamps a signal passing from a first, powered-down power domain to a second, power-on power domain. To reduce leakage current, some of the circuits and devices are connected to a voltage supply of the first or “from” power domain, while other circuits and devices are connected to a voltage supply of the second or “to” power domain.

Abstract: An isolation cell clamps a signal passing from a first, powered-down power domain to a second, power-on power domain. To reduce leakage current, some of the circuits and devices are connected to a voltage supply of the first or “from” power domain, while other circuits and devices are connected to a voltage supply of the second or “to” power domain.

Abstract: A method for configuring level shifter spare cells includes providing a power rail connectable to a corresponding power domain, and providing a spare cell including a level shifter circuit. The level shifter circuit has first and second terminals that are connectable to the power rail, and the first and second terminals are floating with respect to the power rail.

Abstract: A railway vehicle and an axle box structure thereof are provided. The axle box structure includes an axle box body and a positioning node mounting seat. The axle box body includes an upper box body and a lower box body. The upper box body and the lower box body enclose a bearing mounting chamber together. A lower end surface of the upper box body has a convex portion, and an upper end surface of the lower box body has a concave portion. A first semi-circular hole and a second semi-circular hole are respectively provided at the middle of the upper box body and the lower box body. If the convex portion and the concave portion are assembled in a sealed coupling manner, the bearing mounting chamber is formed by the first semi-circular hole and the second semi-circular hole.

Abstract: A railway vehicle and an axle box structure thereof are provided. The axle box structure includes an axle box body and a positioning node mounting seat. The axle box body includes an upper box body and a lower box body. The upper box body and the lower box body enclose a bearing mounting chamber together. A lower end surface of the upper box body has a convex portion, and an upper end surface of the lower box body has a concave portion. A first semi-circular hole and a second semi-circular hole are respectively provided at the middle of the upper box body and the lower box body. If the convex portion and the concave portion are assembled in a sealed coupling manner, the bearing mounting chamber is formed by the first semi-circular hole and the second semi-circular hole.