Amazon.com

About: Amazon.com is a company organization based out in Seattle, Washington, United States. It is known for research contribution in the topics: Service (business) & Service provider. The organization has 13363 authors who have published 17317 publications receiving 266589 citations.

Outcome-oriented dialogs on a speech recognition platform

Abstract: A speech recognition platform configured to receive an audio signal that includes speech from a user and perform automatic speech recognition (ASR) on the audio signal to identify ASR results. The platform may identify: (i) a domain of a voice command within the speech based on the ASR results and based on context information associated with the speech or the user, and (ii) an intent of the voice command. In response to identifying the intent, the platform may perform multiple actions corresponding to this intent. The platform may select a target action to perform, and may engage in a back-and-forth dialog to obtain information for completing the target action. The action may include streaming audio to the device, setting a reminder for the user, purchasing an item on behalf of the user, making a reservation for the user or launching an application for the user.

Handheld electronic book reader device having dual displays

Abstract: A handheld electronic book device is configured with dual displays. The device includes a first display for presenting visible representations of textual or graphic content related to the electronic book. The device also includes a second display positioned alongside the first display. The second display includes a plurality of graphic elements that correspond to portions of the first display. Also, the second display is responsive to user input to one of the graphic elements to perform an action on the content that is shown in the portion of the first display that corresponds to the one element.

The Endogeneity of Union Status: An Empirical Test

Abstract: An unsettled issue in the literature relating to the relative wage effect of unions is the appropriate treatment of union status in a wage determination model. In the context of a three-equation model determining union membership and union- and nonunion-sector wage rates, this paper presents an instrumental variables (IV) procedure for estimating the parameters of the wage equations and a test of the exogeneity of union status using the Hausman specification test. An advantage of our IV procedure in comparison to the widely used inverse Mill's ratio procedure is that our procedure is a distribution-free estimator, whereas the inverse Mill's ratio estimator hinges in the assumption that the error term of the choice equation is normally distributed. Using data for a sample of middle-aged white workers, we estimate the parameters of the union and nonunion wage equations with both procedures. On the key question of the endogeneity of union status, the Hausman test decisively rejects the null hypothesis of exogeneity. The inverse Mill's ratio procedure, in contrast, provides coefficient estimates on the selectivity terms that fail to indicate evidence of sample selectivity in either sector.

d-SNE: Domain Adaptation Using Stochastic Neighborhood Embedding

Abstract: On the one hand, deep neural networks are effective in learning large datasets. On the other, they are inefficient with their data usage. They often require copious amount of labeled-data to train their scads of parameters. Training larger and deeper networks is hard without appropriate regularization, particularly while using a small dataset. Laterally, collecting well-annotated data is expensive, time-consuming and often infeasible. A popular way to regularize these networks is to simply train the network with more data from an alternate representative dataset. This can lead to adverse effects if the statistics of the representative dataset are dissimilar to our target.This predicament is due to the problem of domain shift. Data from a shifted domain might not produce bespoke features when a feature extractor from the representative domain is used. Several techniques of domain adaptation have been proposed in the past to solve this problem. In this paper, we propose a new technique (d-SNE) of domain adaptation that cleverly uses stochastic neighborhood embedding techniques and a novel modified-Hausdorff distance. The proposed technique is learnable end-to-end and is therefore, ideally suited to train neural networks. Extensive experiments demonstrate that d-SNE outperforms the current states-of-the-art and is robust to the variances in different datasets, even in the one-shot and semi-supervised learning settings. d-SNE also demonstrates the ability to generalize to multiple domains concurrently.

Directed audio for speech recognition

Abstract: Techniques are described for selecting audio from locations that are most likely to be sources of spoken commands or words. Directional audio signals are generated to emphasize sounds from different regions of an environment. The directional audio signals are processed by an automated speech recognizer to generate recognition confidence values corresponding to each of the different regions, and the region resulting in the highest recognition confidence value is selected as the region most likely to contain a user who is speaking commands.