公开(公告)号：US11488581B1

公开(公告)日：2022-11-01

申请号：US16706628

申请日：2019-12-06

Applicant: Amazon Technologies, Inc.

Inventor： Shlomi Chovel ,  Adriano Devillaine ,  Omer Shabtai Jakobinsky ,  Colin Zhen De Kho ,  Kawshik Karur Rangaraju ,  Ajay Soni ,  Yochai Zvik ,  Yunqiang Zhu

IPC: G10L15/06 ,  G10L15/187 ,  G10L15/22

Abstract: 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.

公开(公告)号：US20220300560A1

公开(公告)日：2022-09-22

申请号：US17249933

申请日：2021-03-18

Applicant: Amazon Technologies, Inc.

Inventor： Simone Filice ,  Ajay Soni ,  Omer Shabtai Jakobinsky ,  Giuseppe Castellucci ,  Anupama Kumari ,  Vivek Sarthi ,  Oleg Rokhlenko

IPC: G06F16/903 ,  G10L15/22 ,  G06N20/00

Abstract: Contextual data corresponding to previous search requests of a service provider's electronic catalog can be used to resolve voice-input search requests and present search results. Contextual data includes the previous search request that is input to a machine learning algorithm along with a present search request. The machine learning algorithm generates a score indicative of whether the present search request is a refinement of the previous search or a new search request. Once the search request is classified as a refinement or a new search, the search is processed to provide search results including available items from the service provider matching the search request.

公开(公告)号：US11302312B1

公开(公告)日：2022-04-12

申请号：US16585988

申请日：2019-09-27

Applicant: Amazon Technologies, Inc.

Inventor： Ajay Soni ,  Xi Chen ,  Jingqian Zhao ,  Liu Yang ,  Prathap Ramachandra ,  Ruiqi Luo

IPC: G10L15/07 ,  G10L15/06 ,  G10L15/18

Abstract: 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.

公开(公告)号：US11222630B1

公开(公告)日：2022-01-11

申请号：US16576115

申请日：2019-09-19

Applicant: Amazon Technologies, Inc.

Inventor： Ajay Soni ,  Jingqian Zhao ,  Ruiqi Luo ,  Adam Kalman ,  Prathap Ramachandra ,  Liu Yang ,  Simone Filice ,  Ponnu Jacob ,  Amitpal Singh Bhutani

IPC: G10L15/22 ,  G10L15/187 ,  G10L15/16

Abstract: 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.