Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for performing machine learning tasks. One method includes receiving (i) a model input, and (ii) data identifying a first machine learning task to be performed on the model input to generate a first type of model output for the model input; augmenting the model input with an identifier for the first machine learning task to generate an augmented model input; and processing the augmented model input using a machine learning model, wherein the machine learning model has been trained on training data to perform a plurality of machine learning tasks including the first machine learning task, and wherein the machine learning model has been configured through training to process the augmented model input to generate a machine learning model output of the first type for the model input.

Abstract: A speaker sub-assembly has at least one active segment base segment, at least one batch segment, and a passive end segment. In one or more embodiments, speaker sub-assemblies with batched cones may be attached to each other by way of an adapter segment to create a dual-core speaker with batched cones. In one or more embodiments, speaker sub-assemblies with batched cones may be assembled in parallel, or linearly end-to-end.

Abstract: Methods, systems, and computer program products for generating, from an input character sequence, an output sequence of audio data representing the input character sequence. The output sequence of audio data includes a respective audio output sample for each of a number of time steps. One example method includes, for each of the time steps: generating a mel-frequency spectrogram for the time step by processing a representation of a respective portion of the input character sequence using a decoder neural network; generating a probability distribution over a plurality of possible audio output samples for the time step by processing the mel-frequency spectrogram for the time step using a vocoder neural network; and selecting the audio output sample for the time step from the possible audio output samples in accordance with the probability distribution.

Abstract: A ceramic matrix composite (CMC) component and method of forming the CMC component with multiple layers of impregnated matrix fibers, at least one of the multiple layers including at least one CMC prepreg formed from a plurality of twisted tows, twisted through at least one turn/meter.

Abstract: Polymer compositions comprising stiff and tough star shaped styrene butadiene block-copolymers A1 and A2 can be used for making shrink films. Block copolymer A2 preferably has the structure with hard blocks Se and Si, hard random copolymer blocks (B/S)Ae, soft random copolymer blocks (B/S)B coupled by a coupling agent X.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for machine translation using neural networks. In some implementations, a text in one language is translated into a second language using a neural network model. The model can include an encoder neural network comprising a plurality of bidirectional recurrent neural network layers. The encoding vectors are processed using a multi-headed attention module configured to generate multiple attention context vectors for each encoding vector. A decoder neural network generates a sequence of decoder output vectors using the attention context vectors. The decoder output vectors can represent distributions over various language elements of the second language, allowing a translation of the text into the second language to be determined based on the sequence of decoder output vectors.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for machine translation using neural networks. In some implementations, a text in one language is translated into a second language using a neural network model. The model can include an encoder neural network comprising a plurality of bidirectional recurrent neural network layers. The encoding vectors are processed using a multi-headed attention module configured to generate multiple attention context vectors for each encoding vector. A decoder neural network generates a sequence of decoder output vectors using the attention context vectors. The decoder output vectors can represent distributions over various language elements of the second language, allowing a translation of the text into the second language to be determined based on the sequence of decoder output vectors.

Abstract: A seating furniture item with an adjustable-inclination seat part, a back part, a chassis and an at least two-piece leg support connected pivotably on the seat part and/or on the chassis, wherein the parts of the leg support can be displaced relative to one another during pivoting. A part of the leg support is formed by way of a telescopic leg support element which is retracted when the leg support is pivoted in and is extended in order to increase the length of the leg support when the leg support is pivoted out. By way of a first part stroke, a linear actuator brings about a displacement of the seat part relative to the chassis and a simultaneous inclination adjustment of the seat part, wherein no pivoting out or in of the leg support relative to the seat part takes place during this first part stroke of the linear actuator.

Abstract: A piece of seating furniture having a seat part, a back part, a chassis, and an at least two-part leg rest pivotably. The parts of the leg rest can be slid relative to each other during the pivoting. A part of the leg rest is formed by a telescoping leg-rest element which is retracted when the leg rest is pivoted in and is extended, in order to increase the length of the leg rest, when the leg rest is pivoted out. A first articulated lever and a second articulated lever are articulated to the seat part, a third articulated lever and a fourth articulated lever are each connected to a fifth articulated lever. A coupling lever and a control lever are articulated to the first articulated lever and/or the second articulated lever and/or to the third articulated lever and/or the fourth articulated lever.

Abstract: Seating furniture includes at least a first and a second linear actuator to adjust the seat part, the foot part, and the back part between the seating and the reclining position. The first linear actuator generates a linear stroke causing the back part to be adjusted relative to the seat part. The second linear actuator generates a linear stroke having at least a first and second partial stroke, The first partial stroke causes, during adjustment of the seating furniture from the seating position into the reclining position, the seat part to slide relative to the frame towards the front, and simultaneously the inclination of the seat part to be adjusted, there being no extension or retraction of the foot part relative to the seat part during this first partial stroke of the second linear actuator. The second linear actuator causes the foot part to extend relative to the seat part by a second partial stroke following the first partial stroke.

Abstract: The present disclosure relates to a microporous hollow fiber filter membrane having a large inner diameter and a thin wall. The fiber can be used for sterile filtration of liquids or removal of particles from liquids. The disclosure further relates to a method for producing the membrane and a filter device comprising the membrane.

Abstract: Polymer compositions comprising stiff and tough star shaped styrene butadiene block-copolymers A1 and A2 can be used for making shrink films. Block copolymer A2 preferably has the structure with hard blocks Se and Si, hard random copolymer blocks (B/S)Ae, soft random copolymer blocks (B/S)B coupled by a coupling agent X.

Abstract: An item of seating furniture includes a chassis, a seat part having a pivotable inclination, a back part pivotable relative to the seat part, and a foot part, which is connected to the seat part via a foot-part adjustment mechanism and can be pivoted out. To adjust the item of seating furniture from an upright sitting position into a lying position, the seat part is moved forward relative to the chassis. In the sitting position, the foot part is pivoted in, substantially underneath the seat part. In the lying position, the foot part is pivoted out, in front of the seat part, such that the seat part is extended by the foot part. When being pivoted into a lying position, the back part moves forward, together with the seat part, relative to the chassis, and therefore, in any pivoted position, the back part does not project, or only insignificantly projects, beyond the rear edge of the item of seating furniture.

Abstract: An item of seating furniture includes a chassis, a seat part having a pivotable inclination, a back part pivotable relative to the seat part, and a foot part, which is connected to the seat part via a foot-part adjustment mechanism and can be pivoted out. To adjust the item of seating furniture from an upright sitting position into a lying position, the seat part is moved forward relative to the chassis. In the sitting position, the foot part is pivoted in, substantially underneath the seat part. In the lying position, the foot part is pivoted out, in front of the seat part, such that the seat part is extended by the foot part. When being pivoted into a lying position, the back part moves forward, together with the seat part, relative to the chassis, and therefore, in any pivoted position, the back part does not project, or only insignificantly projects, beyond the rear edge of the item of seating furniture.

Abstract: Seating furniture includes at least a first and a second linear actuator to adjust the seat part, the foot part, and the back part between the seating and the reclining position. The first linear actuator generates a linear stroke causing the back part to be adjusted relative to the seat part. The second linear actuator generates a linear stroke having at least a first and second partial stroke, The first partial stroke causes, during adjustment of the seating furniture from the seating position into the reclining position, the seat part to slide relative to the frame towards the front, and simultaneously the inclination of the seat part to be adjusted, there being no extension or retraction of the foot part relative to the seat part during this first partial stroke of the second linear actuator. The second linear actuator causes the foot part to extend relative to the seat part by a second partial stroke following the first partial stroke.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for machine translation using neural networks. In some implementations, a text in one language is translated into a second language using a neural network model. The model can include an encoder neural network comprising a plurality of bidirectional recurrent neural network layers. The encoding vectors are processed using a multi-headed attention module configured to generate multiple attention context vectors for each encoding vector. A decoder neural network generates a sequence of decoder output vectors using the attention context vectors. The decoder output vectors can represent distributions over various language elements of the second language, allowing a translation of the text into the second language to be determined based on the sequence of decoder output vectors.

Abstract: Provided herein are compositions and formulations that are useful for stabilizing one or more bacteria (e.g., through drying). Methods of stabilizing the one or more bacteria are also disclosed.

Abstract: A loudspeaker arrangement having a first loudspeaker comprising a first sound radiating surface and a first loudspeaker basket, and a second loudspeaker comprising a second sound radiating surface and a second loudspeaker basket. The first loudspeaker and the second loudspeaker are arranged opposite each other in a first direction, a cavity is formed between a front side of the first loudspeaker and a front side of the second loudspeaker, and the first loudspeaker basket is directly coupled to the second loudspeaker basket.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for neural machine translation. The method comprises obtaining a first sequence of words in a source language, generating a modified sequence of words in the source language by inserting a word boundary symbol only at the beginning of each word in the first sequence of words and not at the end of each word, dividing the modified sequence of words into wordpieces using a wordpiece model, generating, from the wordpieces, an input sequence of input tokens for a neural machine translation system; and generating an output sequence of words using the neural machine translation system based on the input sequence of input tokens.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for machine translation using neural networks. In some implementations, a text in one language is translated into a second language using a neural network model. The model can include an encoder neural network comprising a plurality of bidirectional recurrent neural network layers. The encoding vectors are processed using a multi-headed attention module configured to generate multiple attention context vectors for each encoding vector. A decoder neural network generates a sequence of decoder output vectors using the attention context vectors. The decoder output vectors can represent distributions over various language elements of the second language, allowing a translation of the text into the second language to be determined based on the sequence of decoder output vectors.