Abstract: A system for estimating the location of a stationary or moving sound source includes multiple microphones, which need not be physically aligned in a linear array or a regular geometric pattern in a given environment, an auralizer that generates auralized multi-channel signals based at least on array-related transfer functions and room impulse responses of the microphones as well as signal labels corresponding to the auralized multi-channel signals, a feature extractor that extracts features from the auralized multi-channel signals for efficient processing, and a neural network that can be trained to estimate the location of the sound source based at least on the features extracted from the auralized multi-channel signals and the corresponding signal labels.

Abstract: This specification describes computer-implemented methods and systems. One method includes receiving, by a neural network of a speech recognition system, first data representing a first raw audio signal and second data representing a second raw audio signal. The first raw audio signal and the second raw audio signal describe audio occurring at a same period of time. The method further includes generating, by a spatial filtering layer of the neural network, a spatial filtered output using the first data and the second data, and generating, by a spectral filtering layer of the neural network, a spectral filtered output using the spatial filtered output. Generating the spectral filtered output comprises processing frequency-domain data representing the spatial filtered output. The method still further includes processing, by one or more additional layers of the neural network, the spectral filtered output to predict sub-word units encoded in both the first raw audio signal and the second raw audio signal.

Abstract: A coupler assembly for coupling a work tool to a work machine. The assembly includes a body, a first locking mechanism, and a second locking mechanism. The first locking mechanism is pivotable about a pivot from a first position to a second position and is configured to be coupled to the work tool in the first position and decoupled in the second position. The second locking mechanism includes a pin having a first end and a second end, a tab coupled to the pin proximate the second end, and a spring disposed between the first end and the second end. The second locking mechanism is axially and pivotably movable about an axis such that the second locking mechanism includes a locked position and an unlocked position. As the first locking mechanism moves from the second position to the first position, the second locking mechanism automatically moves to the locked position.

Abstract: A system for estimating the location of a stationary or moving sound source includes multiple microphones, which need not be physically aligned in a linear array or a regular geometric pattern in a given environment, an auralizer that generates auralized multi-channel signals based at least on array-related transfer functions and room impulse responses of the microphones as well as signal labels corresponding to the auralized multi-channel signals, a feature extractor that extracts features from the auralized multi-channel signals for efficient processing, and a neural network that can be trained to estimate the location of the sound source based at least on the features extracted from the auralized multi-channel signals and the corresponding signal labels.

Abstract: Methods, including computer programs encoded on a computer storage medium, for enhancing the processing of audio waveforms for speech recognition using various neural network processing techniques. In one aspect, a method includes: receiving multiple channels of audio data corresponding to an utterance; convolving each of multiple filters, in a time domain, with each of the multiple channels of audio waveform data to generate convolution outputs, wherein the multiple filters have parameters that have been learned during a training process that jointly trains the multiple filters and trains a deep neural network as an acoustic model; combining, for each of the multiple filters, the convolution outputs for the filter for the multiple channels of audio waveform data; inputting the combined convolution outputs to the deep neural network trained jointly with the multiple filters; and providing a transcription for the utterance that is determined.

Abstract: A transfer case for a normally rear wheel drive vehicle having selectively engageable front wheel drive is provided. The transfer case includes a rear housing with a main drive shaft aperture and an aluminum cast front housing threadably connected with the rear housing. The front housing has a main drive shaft aperture axially aligned with the rear housing main drive shaft aperture. The front housing has an integrally formed cylindrical yoke forming an isolation mount for receiving a polymeric isolator allowing for connection of the transfer case with a chassis structure of the vehicle. The cylindrical yoke has a cylindrical preform therein.

Abstract: Methods, including computer programs encoded on a computer storage medium, for enhancing the processing of audio waveforms for speech recognition using various neural network processing techniques. In one aspect, a method includes: receiving multiple channels of audio data corresponding to an utterance; convolving each of multiple filters, in a time domain, with each of the multiple channels of audio waveform data to generate convolution outputs, wherein the multiple filters have parameters that have been learned during a training process that jointly trains the multiple filters and trains a deep neural network as an acoustic model; combining, for each of the multiple filters, the convolution outputs for the filter for the multiple channels of audio waveform data; inputting the combined convolution outputs to the deep neural network trained jointly with the multiple filters; and providing a transcription for the utterance that is determined.

Abstract: A transfer case for a normally rear wheel drive vehicle having selectively engageable front wheel drive is provided. The transfer case includes a rear housing with a main drive shaft aperture and an aluminum cast front housing threadably connected with the rear housing. The front housing has a main drive shaft aperture axially aligned with the rear housing main drive shaft aperture. The front housing has an integrally formed cylindrical yoke forming an isolation mount for receiving a polymeric isolator allowing for connection of the transfer case with a chassis structure of the vehicle. The cylindrical yoke has a cylindrical preform therein.

Abstract: A coupler assembly for coupling a work tool to a work machine. The assembly includes a body, a first locking mechanism, and a second locking mechanism. The first locking mechanism is pivotable about a pivot from a first position to a second position and is configured to be coupled to the work tool in the first position and decoupled in the second position. The second locking mechanism includes a pin having a first end and a second end, a tab coupled to the pin proximate the second end, and a spring disposed between the first end and the second end. The second locking mechanism is axially and pivotably movable about an axis such that the second locking mechanism includes a locked position and an unlocked position. As the first locking mechanism moves from the second position to the first position, the second locking mechanism automatically moves to the locked position.

Abstract: Systems and methods related to concealing mathematical and logical operations in software. Mathematical and logical operations are disguised by replacing them with logically equivalent expressions. Each equivalent expression has at least two expression constants whose values are based on scaling and bias constants assigned to variables in the original mathematical or logical operation. Each of the expression constants may also be based on additive or multiplicative inverses modulo n of the scaling and bias constants. By replacing the original operations with more complex but logically equivalent expressions containing variables that also involve more operations, the true nature of the original operations is disguised.

Abstract: A heating element is held in position by a rigid holder constructed to surround an ultrasonic bonding tool and is itself fastened in a fixed position relative to the bonding tool, such that the rigid holder and the heating element are held in close proximity to the bonding tool without physically contacting the bonding tool. This serves to heat the bonding tool, facilitating adequate bonding without overly heating the integrated circuit assembly and without interfering with the critical ultrasonic properties of the bonding tool and ultrasonic transducer horn.