公开(公告)号：US10777186B1

公开(公告)日：2020-09-15

申请号：US16190047

申请日：2018-11-13

Applicant: Amazon Technologies, Inc.

Inventor： Stefano Stefani ,  Pramod Gurunath ,  Ashish Singh ,  Katrin Kirchoff ,  Deepikaa Suresh ,  Varun Sembium Varadarajan ,  Vasanth Philomin ,  Vikram Sathyanarayana Anbazhagan ,  Pu Paul Zhao ,  Vijit Gupta ,  Ruoyu Huang

IPC: G10L15/00 ,  G10L15/06 ,  G10L25/78 ,  G10L15/30 ,  G10L15/04 ,  G10L15/26 ,  G10L15/183

Abstract: Techniques for streaming real-time automated speech recognition (ASR) are described. A user can stream audio data to a frontend service of the ASR service. The frontend service can establish a bi-directional connection to an audio decoder host to perform ASR on the data stream. The audio decoder host may include a streaming ASR engine which can analyze chunks of the audio data stream using an acoustic model to divide the audio data into words, and a language model to identify sentences made of the words spoken in the audio file. The acoustic model can be trained using short audio sentence data (e.g., on the order of 30 seconds to a few minutes), enabling the transcription service to accurately transcribe short chunks of audio data. The results are then punctuated and normalized. The resulting transcript is then streamed back to the user over the bi-directional connection.

公开(公告)号：US20240331821A1

公开(公告)日：2024-10-03

申请号：US18194350

申请日：2023-03-31

Applicant: Amazon Technologies, Inc.

Inventor： Vijit Gupta ,  Matthew Chih-Hui Chiou ,  Amiya Kishor Chakraborty ,  Anuroop Arora ,  Varun Sembium Varadarajan ,  Sarthak Handa ,  Amit Vithal Sawant ,  Glen Herschel Carpenter ,  Jesse Deng ,  Mohit Narendra Gupta ,  Rohil Bhattarai ,  Samuel Benjamin Schiff ,  Shane Michael McGookey ,  Tianze Zhang

IPC: G16H15/00 ,  G06F40/40 ,  G16H10/60

CPC classification number: G16H15/00 ,  G06F40/40 ,  G16H10/60

Abstract: Systems and methods for performing medical audio summarizing for medical conversations are disclosed. An audio file and meta data for a medical conversation are provided to a medical audio summarization system. A transcription machine learning model is used by the medical audio summarization system to generate a transcript and a natural language processing service of the medical audio summarization system is used to generate a summary of the transcript. The natural language processing service may include at least four machine learning models that identify medical entities in the transcript, identify speaker roles in the transcript, determine sections of the transcript corresponding to the summary, and extract or abstract phrases for the summary. The identified medical entities and speaker roles, determined sections, and extracted or abstracted phrases may then be used to generate the summary.

公开(公告)号：US10013273B1

公开(公告)日：2018-07-03

申请号：US15189151

申请日：2016-06-22

Applicant: AMAZON TECHNOLOGIES, INC.

Inventor： Diwakar Gupta ,  Sean Alexander Monroe ,  Vijit Gupta

IPC: G06F9/455

CPC classification number: G06F9/45558 ,  G06F2009/45562 ,  G06F2009/45575

Abstract: A provider network (and associated methods) permits virtual machine instances to be terminated but imposes some checks to reduce the risk that an instance is unintentionally terminated. In one example, a safety stock threshold is imposed for a group of instances. If terminating instances within that group would cause a violation of the safety stock threshold, the provider network requests confirmation for the instances to be terminated. In another example, a time delay is imposed after receipt of an instance termination request before the instance is actually terminated. During the time delay, the instance is transitioned to an idle state but otherwise persists on its host computer thereby allowing time to revoke the instance termination. In yet another example, the provider network obtains independent confirmation of the identities of all instances to be terminated before such instances can be terminated pursuant to a termination request.