Abstract: A method includes receiving a training dataset that includes one or more spoken training utterances for training an automatic speech recognition (ASR) model. Each spoken training utterance in the training dataset paired with a corresponding transcription and a corresponding target sequence of auxiliary tokens. For each spoken training utterance, the method includes generating a speech recognition hypothesis for a corresponding spoken training utterance, determining a speech recognition loss based on the speech recognition hypothesis and the corresponding transcription, generating a predicted auxiliary token for the corresponding spoken training utterance, and determining an auxiliary task loss based on the predicted auxiliary token and the corresponding target sequence of auxiliary tokens. The method also includes the ASR model jointly on the speech recognition loss and the auxiliary task loss determined for each spoken training utterance.

Abstract: A method includes receiving, by a first computing device (110) and from a second computing device (118), using a short-range wireless communication unit of the first computing device (110), data including state information of the second computing device (118) and identification information for the second computing device (118). The method also includes, responsive to a distance between the first computing device (110) and the second computing device (118) satisfying a threshold distance, determining a particular template graphical user interface from a plurality of template graphical user interfaces. The method also includes determining one or more operations the second computing device (118) is configured to perform. The method further includes outputting a graphical user interface (114) associated with the remote computing device (118).

Abstract: A wireless power transfer apparatus includes a first control system, a first energy transmission system, and a second energy transmission system. The first control system is coupled to the first energy transmission system, and inputs an energy transmission signal to the first energy transmission system. The first energy transmission system is coupled to the second energy transmission system, and inputs some of the energy transmission signals obtained from the first control system into the second energy transmission system. Both the energy transmission signal obtained by the first energy transmission signal from the first control system and an energy transmission signal obtained by the second energy transmission system from the first energy transmission system supply power to a target device.

Abstract: Systems and methods for generating new images for training a machine-learning model are disclosed. Image data is produced regarding an image captured by an image sensor. The image data is altered such that the style of the image (e.g., color, shading, orientation, etc.) is altered. The altered image data is encoded into a first latent space. An image from a database is selected based on its similarity to the altered image and a decoding of the first latent space. Style encodings of the first latent space are extracted to classify a style of the altered image data in a second latent space. New images are then generated utilizing a reconstructor model that combines the two latent spaces. These new images can be used to train an image-recognition model.

Abstract: A power tool is less likely to have lower workability. A power tool includes a motor including a rotor rotatable about a rotation axis extending in a front-rear direction and a stator, an output unit located frontward from the motor and rotatable in response to a rotational force from the rotor, a motor compartment accommodating the motor, a grip extending downward from the motor compartment, a rear rotor bearing supporting a rear portion of the rotor, and a rear cover covering an opening at a rear end of the motor compartment and holding the rear rotor bearing. The motor compartment includes a guide that guides the rear cover in a first circumferential direction and frontward, a front-rear restrictor that restricts front-rear movement of the rear cover, and a circumferential restrictor that restricts circumferential movement of the rear cover.

Abstract: The impact tool is less likely to have a shorter service life. An impact tool includes a motor, a spindle at least partially located frontward from the motor and rotatable by the motor, a hammer surrounding the spindle, and an anvil at least partially located frontward from the spindle and strikable by the hammer in a rotation direction. The spindle includes an internal space extending frontward from an opening in a rear end face of the spindle. The internal space includes a first space having a first inner diameter and connected to the opening, a second space located frontward from the first space and having a second inner diameter smaller than the first inner diameter, and a third space connected to a front end of the second space and having a third inner diameter larger than the second inner diameter. The third space contains a lubricant oil.

Abstract: An automated speech recognition (ASR) model includes a first encoder, a first encoder, a second encoder, and a second decoder. The first encoder receives, as input, a sequence of acoustic frames, and generates, at each of a plurality of output steps, a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The first decoder receives, as input, the first higher order feature representation generated by the first encoder, and generates a first probability distribution over possible speech recognition hypotheses. The second encoder receives, as input, the first higher order feature representation generated by the first encoder, and generates a second higher order feature representation for a corresponding first higher order feature frame. The second decoder receives, as input, the second higher order feature representation generated by the second encoder, and generates a second probability distribution over possible speech recognition hypotheses.

Abstract: An electric ratchet wrench as one example of the ratchet wrench includes an electric motor, a holder that rotatably holds a socket via a one-way clutch mechanism, and a spindle configured to convert a driving force from the motor into a reciprocating rotation motion of the holder. A space is provided between the socket and the holder. The space includes large interval portions and small interval portions. The small interval portions are adjacent to the large interval portions in a rotation direction (lock direction) of the socket and have distances smaller than distances of the large interval portions. Additionally, the one-way clutch mechanism includes columnar locking pins disposed between the spaces and have a diameter with a size equal to or less than the distances of the large interval portions and exceeding the distances of the small interval portions.

Abstract: An impact tool has a mechanical structure to strike an anvil with a striking force appropriate for a task. An impact tool includes a motor, a hammer rotatable by the motor, an anvil that receives a tip tool and is strikable by the hammer in a rotation direction, an elastic member having a front end supporting the hammer and being extendable and contractable in a front-rear direction to apply an elastic force to the hammer toward the anvil, and a positioner supporting a rear end of the elastic member. The positioner changes a support position of the positioner supporting the rear end of the elastic member in the front-rear direction.

Abstract: An electric ratchet wrench as one example of the ratchet wrench includes an electric motor, a holder that rotatably holds a socket via a one-way clutch mechanism, and a spindle configured to convert a driving force from the motor into a reciprocating rotation motion of the holder. A space is provided between the socket and the holder. The space includes large interval portions and small interval portions. The small interval portions are adjacent to the large interval portions in a rotation direction (lock direction) of the socket and have distances smaller than distances of the large interval portions. Additionally, the one-way clutch mechanism includes columnar locking pins disposed between the spaces and have a diameter with a size equal to or less than the distances of the large interval portions and exceeding the distances of the small interval portions.

Abstract: A power tool includes: a motor having a rotor, which rotates about a motor rotational axis extending in a front-rear direction; an output part, which is disposed more forward than the motor and is rotated in response to rotation of the rotor; a housing including: a motor-housing part, which houses the motor; a grip part, which extends downward from the motor-housing part; a battery-holding part, which is connected to a lower-end portion of the grip part; a control circuit board, which is housed in the housing; and an operation-and-display part, which is disposed at a rear portion of the housing and is electrically connected to the control circuit board.

Abstract: A method includes receiving audio data encoding an utterance, processing, using a speech recognition model, the audio data to generate speech recognition scores for speech elements, and determining context scores for the speech elements based on context data indicating a context for the utterance. The method also includes executing, using the speech recognition scores and the context scores, a beam search decoding process to determine one or more candidate transcriptions for the utterance. The method also includes selecting a transcription for the utterance from the one or more candidate transcriptions.

Abstract: In one embodiment, example systems and methods related to a manner of unifying heterogeneous datasets are provided. Multiple heterogeneous datasets containing traffic or driving data are collected. The records of the datasets are combined, and the records in the combined dataset are ordered into a plurality of time series based on timestamps associated with each record. A Bayesian learning method, such as hidden Markov models, is used to identify traffic primitives in the datasets. Each traffic primitive may include several consecutive records in the combined dataset and may correspond to particular driving actions such as turning left or right, stopping, accelerating, etc. The traffic primitives are used to create a traffic primitive index that can be queried by users or researchers for specific records. These records can be used to train or test one or more learning-based algorithms.

Abstract: In one embodiment, example systems and methods related to a manner of optimizing LiDAR sensor placement on autonomous vehicles are provided. A range-of-interest is defined for the autonomous vehicle that includes the distances from which the autonomous vehicle is interested in collecting sensor data. The range-of-interest is segmented into multiple cubes of the same size. For each LiDAR sensor, a shape is determined based on information such as the number of lasers in each LiDAR sensor and the angle associated with each laser. An optimization problem is solved using the determined shape for each LiDAR sensor and the cubes of the range-of-interest to determine the locations to place each LiDAR sensor to maximize the number of cubes that are captured. The optimization problem may further determine the optimal pitch angle and roll angle to use for each LiDAR sensor to maximize the number of cubes that are captured.

Abstract: An electric ratchet wrench as one example of the ratchet wrench includes an electric motor, a holder that rotatably holds a socket via a one-way clutch mechanism, and a spindle configured to convert a driving force from the motor into a reciprocating rotation motion of the holder. A space is provided between the socket and the holder. The space includes large interval portions and small interval portions. The small interval portions are adjacent to the large interval portions in a rotation direction (lock direction) of the socket and have distances smaller than distances of the large interval portions. Additionally, the one-way clutch mechanism includes columnar locking pins disposed between the spaces and have a diameter with a size equal to or less than the distances of the large interval portions and exceeding the distances of the small interval portions.

Abstract: An electric ratchet wrench as one example of the ratchet wrench includes an electric motor, a holder that rotatably holds a socket via a one-way clutch mechanism, and a spindle configured to convert a driving force from the motor into a reciprocating rotation motion of the holder. A space is provided between the socket and the holder. The space includes large interval portions and small interval portions. The small interval portions are adjacent to the large interval portions in a rotation direction (lock direction) of the socket and have distances smaller than distances of the large interval portions. Additionally, the one-way clutch mechanism includes columnar locking pins disposed between the spaces and have a diameter with a size equal to or less than the distances of the large interval portions and exceeding the distances of the small interval portions.

Abstract: The present disclosure provides a method in a data processing system that includes at least one processor and at least one memory. The at least one memory includes instructions executed by the at least one processor to implement a driving encounter recognition system. The method includes receiving information, from one or more sensors coupled to a first vehicle, determining first trajectory information associated with the first vehicle and second trajectory information associated with a second vehicle, extracting a feature vector, providing the feature vector to a trained classifier, the classifier trained using unsupervised learning based on a plurality of feature vectors, and receiving, from the trained classifier, a classification of the current driving encounter in order to facilitate the first vehicle to perform a maneuver based on the current driving encounter.

Abstract: A method includes receiving audio data encoding an utterance, processing, using a speech recognition model, the audio data to generate speech recognition scores for speech elements, and determining context scores for the speech elements based on context data indicating a context for the utterance. The method also includes executing, using the speech recognition scores and the context scores, a beam search decoding process to determine one or more candidate transcriptions for the utterance. The method also includes selecting a transcription for the utterance from the one or more candidate transcriptions.

Abstract: The present invention relates to methods of electron beam lithography using ice resist to fabricate nanostructures on a substrate and, more particularly, to a method of fabricating desired three-dimensional nanostructures on a substrate. The method involves two main strategies: grayscale ice lithography and stacking layered structures. Moreover, these two strategies can be combined in one fabrication process to produce more complex 3D nanostructures.