Apple Inc.

Cisco Technology, Inc.

Facebook, Inc.

Matthias Paulik

Elizabeth Caroline Cranfill

Craig Michael Federighi

Andre Souza Dos Santos

Grant R Paul

Nathan De Vries

Steven D Baker

Ryan S Dixon

Srinivasan Venkatachary

Johnnie B Manzari

Harry Joseph Saddler

Vignesh Jagadeesh

Henry Graham Mason

Laurence Steven Gillick

Alexander Waibel

Marco Zuliani

James Nicholas Jones

Anumita Biswas

Viktor Miladinov

Neil Gareth Crane

Jeffrey A Brasket

Robert A Walker

Vincent M Lane

Sachin S Kajarekar

Gunnar Evermann

Ananth Sankar

Vinay Sharma

Brandon J Corey

Guillaume Borios

Xin Wang

Jason A Skinder

Rushin N Shah

Ashish Garg

Nicholas D Lupinetti

Kellie L Albert

Matthew S Seigel

Shweta Grampurohit

Aparna Khare

Nicolas Zeitlin

Arnaud Legendre

Shaun Eric Williams

Olga Iosifovna Gurevich

Karric Kwong

Pulah J Shah

Thomas Deselaers

Yang Zhao

Matthias Dantone

Kayur Dushyant Patel

Vivek Halder

Udhyakumar Nallasamy

Rongqing Huang

Alex Waibel

Ron C Santos

Ali S Mohamed

Ngoc Hong Pham

Neha Agrawal

Mahesh Krishnamoorthy

Aya Siblini

Sinan Akay

Chen Ye

Zinaida A Pozen

Andrew M Finch

Yun Tang

Francisco Alvaro Munoz

Saravana Kumar Rathinam

Robert A Walker, Ii

Selen Uguroglu

Stephan Peitz

Henry G Mason

Mallika Priya Khullar

Xishuo Liu

Adam Huff Bradford

Guangyu Zhong

Sebastien V Marineau-Mes

Jennifer Pon Chen

Siyang Tang

Amittai Axelrod

Rajen Chatterjee

Systems and processes for providing user-specific acoustic models are provided. In accordance with one example, a method includes, at an electronic device having one or more processors, receiving a plurality of speech inputs, each of the speech inputs associated with a same user of the electronic device; providing each of the plurality of speech inputs to a user-independent acoustic model, the user-independent acoustic model providing a plurality of speech results based on the plurality of speech inputs; initiating a user-specific acoustic model on the electronic device; and adjusting the user-specific acoustic model based on the plurality of speech inputs and the plurality of speech results.

User-specific acoustic models

Aspects of subject technology provide systems and methods for simultaneously spell-correcting and completing partial search queries being entered by a user on the user's electronic device. An apparatus such as a computing device may receive partial search queries from the user's electronic device as each character of the partial search query is entered by the user. The apparatus may utilize a machine-learning model to generate suggested queries that include spelling-corrected versions of the received partial query, query completion suggestions for the partial query, and/or spelling-corrected completion suggestions for the partial query.

Correction and completion of search queries

Systems and processes for operating an electronic device to train a machine-learning translation system are described. In one process, a first set of training data is obtained. The first set of training data includes at least one payload in a first language and a translation of the at least one payload in a second language. The process further includes obtaining one or more templates for adapting the at least one payload; adapting the at least one payload using the one or more templates to generate at least one adapted payload formulated as a translation request; generating a second set of training data based on the at least one adapted payload; and training the machine-learning translation system using the second set of training data.

Implicit identification of translation payload with neural machine translation

Systems and processes for performing a task with a digital assistant are provided. In accordance with one example, a method includes, at an electronic device having one or more processors, receiving a natural-language input; determining, based on the natural-language input, a first task and first usefulness score associated with the first task; receiving, from another electronic device, a second task and second usefulness score associated with the second task; determining whether the first usefulness score is higher than the second usefulness score; in accordance with a determination that the first usefulness score is higher than the second usefulness score: performing the first task determined by the electronic device; and providing an output indicating whether the first task has been performed; and in accordance with a determination that the second usefulness score is higher than the first usefulness score: performing the second task received from the another electronic device; and providing an output indicating whether the second task has been performed.

Offline personal assistant

An iterative language translation system includes multiple communicatively connected statistical speech translation systems. The system includes an automatic speech recognition component adapted to recognize spoken language in a source language and to create a source language hypothesis. A machine translation component is adapted to translate the source language hypothesis into a target language. The system also includes a second automatic speech recognition component and second machine translation component. The translation results are used to adapt the automatic speech recognition components and the language hypotheses are used to adapt the machine translation components.

Training statistical speech translation systems from speech

Speech recognition is performed on a received utterance to determine a plurality of candidate text representations of the utterance, including a primary text representation and one or more alternative text representations. Natural language processing is performed on the primary text representation to determine a plurality of candidate actionable intents, including a primary actionable intent and one or more alternative actionable intents. A result is determined based on the primary actionable intent. The result is provided to the user. A recognition correction trigger is detected. In response to detecting the recognition correction trigger, a set of alternative intent affordances and a set of alternative text affordances are concurrently displayed.

User interface for correcting recognition errors

Systems and processes for processing speech in a digital assistant are provided. In one example process, a first speech input can be received from a user. The first speech input can be processed using a first automatic speech recognition system to produce a first recognition result. An input indicative of a potential error in the first recognition result can be received. The input can be used to improve the first recognition result. For example, the input can include a second speech input that is a repetition of the first speech input. The second speech input can be processed using a second automatic speech recognition system to produce a second recognition result.

Automatic speech recognition based on user feedback

Systems and processes are disclosed for discovering trending terms in automatic speech recognition. Candidate terms (e.g., words, phrases, etc.) not yet found in a speech recognizer vocabulary or having low language model probability can be identified based on trending usage in a variety of electronic data sources (e.g., social network feeds, news sources, search queries, etc.). When candidate terms are identified, archives of live or recent speech traffic can be searched to determine whether users are uttering the candidate terms in dictation or speech requests. Such searching can be done using open vocabulary spoken term detection to find phonetic matches in the audio archives. As the candidate terms are found in the speech traffic, notifications can be generated that identify the candidate terms, provide relevant usage statistics, identify the context in which the terms are used, and the like.

Method and apparatus for discovering trending terms in speech requests

Systems and processes for automatic accent detection are provided. In accordance with one example, a method includes, at an electronic device with one or more processors and memory, receiving a user input, determining a first similarity between a representation of the user input and a first acoustic model of a plurality of acoustic models, and determining a second similarity between the representation of the user input and a second acoustic model of the plurality of acoustic models. The method further includes determining whether the first similarity is greater than the second similarity. In accordance with a determination that the first similarity is greater than the second similarity, the first acoustic model may be selected; and in accordance with a determination that the first similarity is not greater than the second similarity, the second acoustic model may be selected.

Automatic accent detection using acoustic models

The present disclosure generally relates to context-based endpoint detection in user speech input. A method for identifying an endpoint of a spoken request by a user may include receiving user input of natural language speech including one or more words; identifying at least one context associated with the user input; generating a probability, based on the at least one context associated with the user input, that a location in the user input is an endpoint; determining whether the probability is greater than a threshold; and in accordance with a determination that the probability is greater than the threshold, identifying the location in the user input as the endpoint.

Context-based endpoint detection

Systems and processes for evaluating embedded personalized systems are provided. In one example process, instructions that define an experiment associated with a personalized speech recognition system can be received. The instructions can define one or more experimental parameters. In accordance with the received instructions, a second personalized speech recognition system can be generated based on the personalized speech recognition system and the one or more experimental parameters. Additionally, the plurality of user speech samples can be processed using the second personalized speech recognition system to generate a plurality of speech recognition results and a plurality of accuracy scores corresponding to the plurality of speech recognition results. Second instructions can be received based on the plurality of accuracy scores. In accordance with the second instructions, the second speech recognition system can be activated.

Privacy preserving distributed evaluation framework for embedded personalized systems

Growing community of inventors

San Jose, CA, United States of America

Matthias Paulik