Lucent Technologies Inc.

Agere Systems Inc.

Agere Systems Guardian Corp.

Adesto Technologies Corporation

Applied Materials, Inc.

Spansion Llc.

Botbotbotbot Inc.

Apple Inc.

Omnivision Technologies, Inc.

Monterey Research, LLC

Adesto Technologies France Sarl

Yi Ma

Pradip Kumar Roy

Khaled Ahmed

Chakravarthy Gopalan

Yih-Feng Chyan

Shenqing Fang

Robert B Ogle

David Charles Brady

Sailesh Mansinh Merchant

Christopher Sean Olsen

Jeffrey Allan Shields

Abhilash J Mayur

Isik C Kizilyalli

Wei Ti Lee

Chung Wai Leung

Shreyas Kher

Antonio R Gallo

Janet Wang

Antoine Raux

Michael A Laughery

Duli Mao

Gang Chen

Philip Allan Kraus

Dyson Hsinchih Tai

Majeed Foad

Venkatesh P Gopinath

Mark Yam

Maitreyee Mahajani

Yi-Chiau Huang

Haifan Liang

Jallepally Ravi

Rinji Sugino

Shulin Wang

Foroozan Sarah Koushan

Yuanwei Zheng

Gary Eugene Miner

Allen Yen

Stephen Carl Kuehne

Kevin Laughton Cunningham

Toi Yue Becky Leung

Minseok Oh

Robert C McIntosh

Antoine R Raux

Tejal Goyani

Jane Qian Liu

Jean Ling Lee

Paul Yang

Felicia Wong

Lionel Cooper Kimerling

Kuo-Tung Chang

Pravin K Narwankar

Yuan Chen

Jean Y Yang

Chun-Ting Liu

Kurt George Steiner

Youseok Suh

Brittin C Kane

Alan Sangone Chen

Clifford Alan King

John Dinh

Inkuk Kang

Mark D Kellam

Yuanning Chen

Michael Scott Carroll

John Sanchez

John Ross Jameson

Marco Mastrapasqua

Shahriar Moinian

Chunchieh Huang

Gang Xue

Martin Laurence Green

Kuei-Chang Tsai

John Michael Hergenrother

Steven J Hillenius

Donald Paul Monroe

Kin Ping Cheung

Gregg Sumio Higashi

Kwok K Ng

Scott Francis Shive

Michel Ranjit Frei

Eunha Kim

Philippe Blanchard

John Ross Jameson, Iii

Kevin D Pitolin

Timothy Scott Campbell

Minesh Amrat Patel

William John Nagy

Huili Shao

Felicia W Edwards

Kevin Yun-kang Wu

Timothy W Weidman

Hongzong Chew

Robert Ys Huang

Joseph Ashley Taylor

Ming Kwon

Edith Yang

Demi Nguyen

Hooi Peng Low

Kean Syn Cheah

Melissa M Brown

Systems and processes for operating a digital assistant are provided. An example method includes, at an electronic device having one or more processors and memory, detecting invocation of a digital assistant; determining, using a reference resolution service, a set of possible entities; receiving a user utterance including an ambiguous reference; determining based on the user utterance and the list of possible entities, a candidate interpretation including a preliminary set of entities corresponding to the ambiguous reference; determining, with the reference resolution service and based on the candidate interpretation including the preliminary set of entities corresponding to the ambiguous reference, an entity corresponding to the ambiguous reference; and performing, based on the candidate interpretation and the entity corresponding to the ambiguous reference, a task associated with the user utterance.

Reference resolution during natural language processing

Systems and processes for operating an intelligent automated assistant are provided. An example process includes, at an electronic device having one or more processors and memory: performing a first task specified in a first user speech input; receiving a second user speech input; and in accordance with a determination that the second user speech input includes a modification to the first task, performing a second task, wherein performance of the second task modifies at least a portion of the performance of the first task.

Task modification after task initiation

A method to fabricate a non-planar memory device including forming a multi-layer silicon nitride structure substantially perpendicular to a top surface of the substrate. There may be multiple non-stoichiometric silicon nitride layers, each including a different or same silicon richness value from one another.

Varied silicon richness silicon nitride formation

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for secure data transmission using natural language. One of the methods includes: obtaining sensitive information for a user; obtaining a natural language key for the user, wherein the natural language key for the user includes one or more natural language tokens; generating decoding data for the sensitive information for the user, wherein generating the decoding data comprises: for each place in the sensitive information for the user: assigning a respective one of the natural language tokens in the natural language key for the user to the value at the place, and generating one or more respective dummy natural language tokens for each value of the respective set of possible values for the place other than the value at the place; and providing the decoding data for use in decoding the natural language key into the sensitive information.

Secure data transmission using natural language

An image sensor includes a plurality of photodiodes disposed in a semiconductor material, and a through-semiconductor-via coupled to a negative voltage source. Deep trench isolation structures are disposed between individual photodiodes in the plurality of photodiodes to electrically and optically isolate the individual photodiodes. The deep trench isolation structures include a conductive material coupled to the through-semiconductor-via, and a dielectric material disposed on sidewalls of the deep trench isolation structures between the semiconductor material and the conductive material.

Biased deep trench isolation

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating additional training data for a natural language understanding engine. One of the methods includes: obtaining data identifying (i) a first input conversational turn and (ii) a first annotation, determining that the first annotation accurately characterized the first input conversational turn, determining that the natural language understanding engine is likely to generate inaccurate annotations of other conversational turns that are similar to the first input conversational turn, in response to the determining, obtaining one or more first paraphrases of the first input conversational turn; and generating, for each of the one or more first paraphrases, a respective first training example that identifies the first annotation as the correct annotation for the first paraphrase; and training the natural language understanding engine on at least the first training examples.

Generating additional training data for a natural language understanding engine

In one embodiment of the present invention, a resistive switching device includes a first electrode disposed over a substrate and coupled to a first potential node, a switching layer disposed over the first electrode, a conductive amorphous layer disposed over the switching layer, and a second electrode disposed on the conductive amorphous layer and coupled to a second potential node.

Resistive switching devices having a switching layer and an intermediate electrode layer and methods of formation thereof

Programmable impedance elements structures, devices and methods are disclosed. Methods can include: forming a first electrode layer within an electrode opening that extends through a cap layer; planarizing to expose a top of the cap layer; cleaning the exposed top surface of the cap layer to remove residual species from previous process steps. Additional methods can include forming at least a base ion conductor layer having an active metal formed therein that may ion conduct within the ion conductor layer; and forming an inhibitor material that mitigates agglomeration of the active metal within the base ion conductor layer as compared to the active metal alone. Programmable impedance elements and/or devices can have switching material and electrodes parallel to both bottoms and sides of a cell opening formed in a cell dielectric. Other embodiments can include an ion conductor layer having an alloy of an active metal, or two ion conductor layers in contact with an active electrode.

Programmable impedance elements and devices that include such elements

A memory device can include a plurality of memory cells formed over a substrate, each memory cell including a tunnel access device that enables current flow in at least one direction predominantly due to tunneling, and a storage element programmable between different impedance states by a reduction-oxidation reaction within at least one memory layer formed between two electrodes; wherein the tunneling access device and programmable impedance element are vertically stacked over one another.

Memory cells with vertically integrated tunnel access device and programmable impedance element

A method of forming a conductive bridging memory cell can include forming an active electrode layer above a barrier layer formed on a lower conductive layer; forming at least one ion conductor layer over an active electrode layer; incorporating conductive ions into the ion conductor layer to create a switch memory layer that changes impedance in response to an electric field; and the active electrode layer is a source of conductive ions for the ion conductor, and the barrier layer substantially prevents a movement of conductive ions therethrough.

Fabrication methods of conducting bridge random access memory (CBRAM) device structures

A memory element can include a first electrode; a second electrode; and a memory material programmable between different resistance states, the memory material disposed between the first electrode and the second electrode and comprising a solid electrolyte with at least one modifier element formed therein; wherein the first electrode is an anode electrode that includes an anode element that is ion conductible in the solid electrolyte, the anode element being different than the modifier element.

Solid electrolyte memory elements with electrode interface for improved performance

A method, in one embodiment, can include forming a tunnel oxide layer on a substrate. In addition, the method can include depositing via atomic layer deposition a first layer of silicon nitride over the tunnel oxide layer. Note that the first layer of silicon nitride includes a first silicon richness. The method can also include depositing via atomic layer deposition a second layer of silicon nitride over the first layer of silicon nitride. The second layer of silicon nitride includes a second silicon richness that is different than the first silicon richness.

Growing community of inventors

Santa Clara, CA, United States of America

Yi Ma