Amazon Technologies, Inc.

Markus Dreyer

Ariya Rastrow

Ankur Gandhe

Anjishnu Kumar

Faisal Ladhak

Lambert Leo Mathias

Björn Hoffmeister

Rohit Prasad

Spyridon Matsoukas

Denis Sergeyevich Filimonov

Arpit Gupta

Nikko Ström

Avani Deshpande

William Clinton Dabney

Boya Yu

Pavankumar Reddy Muddireddy

Julian Chan

Adrian Mark McLeod

Sujith Ravi

Naga Sai Likhitha Patha

Can Liu

Techniques for generating a summary of text-based documents are described. A system may be configured to generate a summary based on context data. The system may receive different types of context data corresponding to a user input. The context data may be converted to a linearized representation so that it can be processed by a decoder along with a source document for which the summary is being generated.

Meaning summarization techniques

The present disclosure describes the generation and use of word embeddings as part of natural language understanding (NLU) processing performed by a natural language processing system. In at least some examples, the word embeddings may be generated from text corpuses including at least text (representing spoken user inputs) output from automatic speech recognition (ASR) processing. In at least some examples, the word embeddings may be generated from text output from ASR processing and natural language text corresponding to one or more Internet webpages.

Word embeddings for natural language processing

A voice user interface (VUI) system use collaborative filtering to expand its own knowledge base. The system is designed to improve the accuracy and performance of the Natural Language Understanding (NLU) processing that underlies VUIs. The system leverages the knowledge of system users to crowdsource new information.

Voice user interface knowledge acquisition system

Neural networks may be used in certain automatic speech recognition systems. To improve performance at these neural networks, the present system converts the lattice into a matrix form, thus maintaining certain information included in the lattice that might otherwise be lost while also placing the lattice in a form that may be manipulated by other components to perform operations such as checking ASR results. The matrix representation of the lattice may be transformed into a vector representation by calculations performed at a recurrent neural network (RNN). By representing the lattice as a vector representation the system may perform additional operations, such as ASR results confirmation.

Lattice decoding and result confirmation using recurrent neural networks

An automatic speech recognition (ASR) system may convert an ASR output lattice into a matrix form, thus maintaining certain information included in the lattice that might otherwise be lost in an N-best list output. The matrix representation of the lattice may be encoded using a recurrent neural network (RNN) to create a vector representation of the lattice. The vector representation may then be used by the system to perform additional operations, such as ASR results confirmation.

Lattice encoding using recurrent neural networks

An approach to extending the recognizable labels of a label recognizer makes use of an encoding of linguistic inputs and label attributes into comparable vectors. The encodings may be determined with artificial neural networks (ANNs) that are jointly trained, and a comparison between the encoding of a sentence input and the encoding of an intent attribute vector may use a fixed function, which does not have to be trained. The encoding of label attributes can generalize permitting adding of a new label via corresponding attributes, thereby avoiding the need to immediately retrain a label recognizer with example inputs.

Extendable label recognition of linguistic input

An automatic speech recognition ('ASR') system produces, for particular users, customized speech recognition results by using data regarding prior interactions of the users with the system. A portion of the ASR system (e.g., a neural-network-based language model) can be trained to produce an encoded representation of a user's interactions with the system based on, e.g., transcriptions of prior utterances made by the user. This user-specific encoded representation of interaction history is then used by the language model to customize ASR processing for the user.

Speech processing with learned representation of user interaction history

A linguist classifier, for instance intent or slot classifier, is updated using data with only partial annotation indicating overall correctness rather that specific correct intent or slot values, which are treated as 'latent' (i.e., unknown) variables. Full annotation of the data is not required. A small amount of fully annotated data may be combined with a substantially larger amount of partially annotated data to update the linguistic classifier. In a specific implementation, the linguistic classifier is a neural network and the weights are trained using a reinforcement learning approach.

Growing community of inventors

Seattle, WA, United States of America

Markus Dreyer