Amazon Technologies, Inc.

Microsoft Technology Licensing, LLC

Ajay Kumar Soni

Prathap Ramachandra

Liu Yang

Jingqian Zhao

Ruiqi Luo

Xi Chelsea Chen

Stephane Taine

Ponnu Jacob

Amitpal Singh Bhutani

Yochai Zvik

Addison Phillips

Adriano Devillaine

Adam Kalman

Yunqiang Zhu Zhu

Omer Shabtai Jakobinsky

Simone Filice

Kuldeep Sharma

Shlomi Chovel

Colin Zhen De Kho

Kawshik Karur Rangaraju

Sarah Giles

Kurt K De Ding

A new approach to automatic speech recognition is disclosed. An example method include receiving a first text representing speech recognition of a phrase spoken by a user, isolating a candidate named entity from within the phrase, receiving a first phonetic representation of the candidate named entity, comparing the first phonetic representation to phonetic representations in a mapping database which map the phonetic representations to words to yield a comparison, based on the comparison, identifying a second phonetic representation in the mapping database that matches a second text in the mapping database to the second phonetic representation and replacing the candidate named entity with the second text. The approach can be used for new brands for which automatic speech recognition error rates are high.

System and method of providing recovery for automatic speech recognition errors for named entities

A new model is introduced into a particular domain that receives a routing of a dialog from a speech processing component. A method associated with the model includes running a set of test utterances through the speech processing component that enables a spoken language dialog with a user to establish a base line score associated with processing for the set of test utterances. The speech processing component determines an intent of the user and routes the spoken language dialog to a network-based domain based on the intent. The method includes establishing an automatic test run of the set of test utterances to obtain a current score and, when a threshold associated with a difference between the current score and the base line score is breached, switching, at the network-based domain, from the false accept detection model to a second model.

Spoken language quality automatic regression detector background

A new model is introduced into a particular domain that receives a routing of a dialog from a speech processing component. The speech processing component is engaged in the dialog with a user and the speech processing component routes the dialog to the particular network-based domain according to a determination by the speech processing component that the user has an intent to perform a task handled by the domain. The model detects, at the domain, whether the user has the proper intent associated with the domain by using the user utterance in its entirety to yield a detection result. When the user does not have the proper intent based on the detection result, the domain drops the user utterance.

Detecting false accepts in a shopping domain for handling a spoken dialog

Systems, methods, and computer-readable media are disclosed for smart code scanning, analysis and reporting. The systems, methods, and computer-readable media described herein may reduce the computational load on the scanner service and improve user experiences. In an example method described herein, a scanning service may receive code for analysis, the code being associated with a first project identifier. The scanner service may scan the code to identify one or more potential issues associated with the code. In turn, the scanner service may provide a list comprising, at least, one or more potential issues for display at a user device. The scanner service may then receive data indicative of a first potential issue being a false positive based on, at least in part, user input. The scanner service may determine, based on the first potential issue being a false positive, a first set of one or more rules for filtering the one or more potential issues, wherein the first set of one or more rules comprises a first rule. The scanner service may store and automatically apply the first set of rules.

Methods and systems for smart code scanning, analysis and reporting

Technologies are generally described for providing an intermediary API for intercepting and processing subscribing requests to a communications server. An intermediary API may intercept a subscribing request from a local or PIC client to receive presence status data from a presentity such as a bot executed on the communications server. The intermediary API may be configured to provide the presence data to a PIC client without consuming the communications server registered subscription limit. The intermediary API may employ a server-supported protocol-specific processing language or module to intercept the subscribing requests, to determine if the request is targeted for one or more bots executed on the communications server, and to determine if the request is for presence data for one or more bots. The intermediary API may determine if the request is non-terminating, and if all criteria are met, may provide presence data to the requesting client.

Intermediary API for providing presence data to requesting clients

Growing community of inventors

Kirkland, WA, United States of America

Ajay Kumar Soni