公开(公告)号：US20250086380A1

公开(公告)日：2025-03-13

申请号：US18957409

申请日：2024-11-22

Applicant: Amazon Technologies, Inc.

Inventor： Monica Lakshmi Sunkara ,  Deepthi Devaiah Devanira ,  Chaitanya Shivade ,  Sravan Babu Bodapati ,  Katrin Kirchhoff ,  Srikanth Ronanki

IPC: G06F40/166 ,  G06F21/62 ,  G06F40/279 ,  G10L15/16 ,  G10L15/22

Abstract: Portions of text data generated from inverse text normalization may be redacted. Text data for redaction may be obtained. One or more inverse text normalization models may be applied to the text data to generate normalized text data. A machine learning model, trained to recognize text for redaction, may be applied to identify portions of the normalized text data for redaction. The identified portions may be redacted and the redacted normalized text provided to a destination.

公开(公告)号：US20250029603A1

公开(公告)日：2025-01-23

申请号：US18356116

申请日：2023-07-20

Applicant: Amazon Technologies, Inc.

Inventor： Karthik Gopalakrishnan ,  Sravan Babu Bodapati ,  Katrin Kirchhoff ,  Sarthak Handa

IPC: G10L15/183 ,  G10L15/06

Abstract: Domain specialty instructions may be generated for performing text analysis tasks. An input text may be received for performing a text analysis task. A domain specialty may be identified for the input text. Specialty domain identifiers may be inserted as part of generating instructions to perform the text analysis task using a pre-trained large language model fine-tuned to a domain that includes multiple domain specialties. The pre-trained large language model may perform the text analysis task on the input text using the generated instructions. A result of the text analysis tsk performed on the input text may be provided.

公开(公告)号：US12205584B1

公开(公告)日：2025-01-21

申请号：US17532986

申请日：2021-11-22

Applicant: Amazon Technologies, Inc.

Inventor： John Baker ,  Anubhav Mishra ,  Bangrui Liu ,  Christopher Michael Hittner ,  Sravan Babu Bodapati ,  Harshal Pimpalkhute ,  Katrin Kirchhoff ,  Anuj Gautam Surana ,  Yilai Su ,  Brandon Louis Mendez ,  Chengshun Zhang

IPC: G10L15/22 ,  G10L13/027 ,  G10L15/08

Abstract: A set of alternative vocal input styles for specifying a parameter of a dialog-driven application is determined. During execution of the application, an audio prompt requesting input in one of the styles is presented. A value of the parameter is determined by applying a collection of analysis tools to vocal input obtained after the prompt is presented. A task of the application is initiated using the value.

公开(公告)号：US12136413B1

公开(公告)日：2024-11-05

申请号：US17710762

申请日：2022-03-31

Applicant: Amazon Technologies, Inc.

Inventor： Saket Dingliwal ,  Sravan Babu Bodapati ,  Katrin Kirchhoff ,  Ankur Gandhe ,  Anubhav Mishra ,  John Baker ,  Ashish Vishwanath Shenoy ,  Ravi Teja Gadde

IPC: G06F40/40 ,  G10L15/06 ,  G10L15/183

Abstract: Domain-specific parameters may be used for tuning speech processing. A pre-trained transformer-based language model may train domain-specific parameters using domain-specific unlabeled text data. This domain-specific parameters can then be appended to candidate texts produced by a speech model on received speech data and input to the transformer-based language model to score the candidate texts. The scores of the candidate texts determined using the pre-trained transformer-based language model can then be used to select a candidate text for further speech processing.

公开(公告)号：US12223259B1

公开(公告)日：2025-02-11

申请号：US16587800

申请日：2019-09-30

Applicant: Amazon Technologies, Inc.

Inventor： Varun Sembium Varadarajan ,  Sravan Babu Bodapati ,  Deepthi Devaiah Devanira ,  Pu Paul Zhao ,  Katrin Kirchhoff ,  Yue Yang

IPC: G06F40/166 ,  G06F18/214 ,  G06F21/62 ,  G06F40/279 ,  G06F40/30 ,  G06N20/00

Abstract: Techniques for managing access to sensitive data in transcriptions are described. A method for managing access to sensitive data in transcriptions may include receiving a request to generate a redacted transcript of content, obtaining a transcript of the content, sending at least a portion of the transcript to a model endpoint to identify sensitive entities in the transcript, receiving an inference response identifying one or more sensitive entities in the transcript, and generating the redacted transcript based at least one the transcript and the inference response.

公开(公告)号：US12198681B1

公开(公告)日：2025-01-14

申请号：US17937297

申请日：2022-09-30

Applicant: Amazon Technologies, Inc.

Inventor： Monica Lakshmi Sunkara ,  Srikanth Ronanki ,  Sravan Babu Bodapati ,  Jeffrey John Farris ,  Katrin Kirchhoff ,  Vivek Govindan ,  Yide Zou ,  Mohit Narendra Gupta ,  Silviu Mihai Burz

IPC: G10L15/16 ,  G10L15/30 ,  G10L25/51

Abstract: Techniques for personalized batch and streaming speech-to-text transcription of audio reduce the error rate of automatic speech recognition (ASR) systems in transcribing rare and out-of-vocabulary words. The techniques achieve personalization of connectionist temporal classification (CT) models by using adaptive boosting to perform biasing at the level of sub-words. In addition to boosting, the techniques encompass a phone alignment network to bias sub-word predictions towards rare long-tail words and out-of-vocabulary words. A technical benefit of the techniques is that the accuracy of speech-to-text transcription of rare and out-of-vocabulary words in a custom vocabulary by automatic speech recognition (ASR) system can be improved without having to train the ASR system on the custom vocabulary. Instead, the techniques allow the same ASR system trained on a base vocabulary to realize the accuracy improvements for different custom vocabularies spanning different domains.

公开(公告)号：US11580965B1

公开(公告)日：2023-02-14

申请号：US16938783

申请日：2020-07-24

Applicant: Amazon Technologies, Inc.

Inventor： Monica Lakshmi Sunkara ,  Srikanth Ronanki ,  Dhanush Bekal Kannangola ,  Sravan Babu Bodapati ,  Katrin Kirchhoff

IPC: G10L15/19 ,  G06N3/049 ,  G10L15/26 ,  G10L15/06

Abstract: Techniques for predicting punctuation and casing using multimodal fusion are described. An exemplary method includes processing generated text by: tokenizing the generated text into sub-words, and generating a sequence of lexical features for the sub-words using a pre-trained lexical encoder; processing audio of the audio by: generating a sequence of frame level acoustic embeddings using a pre-trained acoustic encoder on the audio, and generating task specific embeddings from the frame level acoustic embeddings; performing multimodal fusion of the sub-word level acoustic embeddings and the sequence of lexical features by: aligning the task specific embeddings to the sequence of lexical features, and combining the sequence of lexical features and aligned acoustic sequence; predicting punctuation and casing from the combined sequence of lexical features and aligned acoustic sequence; concatenating the sub-words of the text, and applying the predicted punctuation and casing; and outputting text having the predicted punctuation and casing.

公开(公告)号：US20250005282A1

公开(公告)日：2025-01-02

申请号：US18344764

申请日：2023-06-29

Applicant: Amazon Technologies, Inc.

Inventor： John Colton Moriarty ,  Saket Dingliwal ,  Karthik Gopalakrishnan ,  Sravan Babu Bodapati ,  Katrin Kirchhoff ,  Lei Xu

IPC: G06F40/284 ,  G06F16/34

Abstract: Domain specialty instructions may be generated for performing text analysis tasks. An input text may be received for performing a text analysis task. One or more domain entities may be extracted from the input text using a machine learning model trained to recognize entities of a domain in a given text. The one or more domain entities may be inserted as part of generating instructions to perform the text analysis task using a pre-trained machine learning model fine-tuned to the domain. The pre-trained machine learning model may be caused to perform the text analysis task using the generated instructions and a result of the text analysis task may be provided.

公开(公告)号：US20250005063A1

公开(公告)日：2025-01-02

申请号：US18344739

申请日：2023-06-29

Applicant: Amazon Technologies, Inc.

Inventor： Devang Kulshreshtha ,  Saket Dingliwal ,  Sravan Babu Bodapati ,  Katrin Kirchhoff ,  Sarthak Handa

IPC: G06F16/34 ,  G06F40/169 ,  G06F40/40

Abstract: Pairs of text collections are obtained. An individual pair comprises (a) a source text collection which includes a first group of text sequences and (b) an annotated analysis result of the source text collection, comprising a second group of text sequences and a set of evidence mappings generated by an evidence mapping model. An evidence mapping indicates, for a particular text sequence of the second group, another text sequence of the first group which provides evidence for the particular text sequence. A quality metric of the model is obtained using an automated evaluation methodology in which a question is generated from the particular text sequence, and an analysis of a pair of answers (including 10 an answer generated using an evidence mapping) to the question is performed. The quality metric is provided via a programmatic interface.

公开(公告)号：US12182498B1

公开(公告)日：2024-12-31

申请号：US17810302

申请日：2022-06-30

Applicant: Amazon Technologies, Inc.

Inventor： Monica Lakshmi Sunkara ,  Deepthi Devaiah Devanira ,  Chaitanya Shivade ,  Sravan Babu Bodapati ,  Katrin Kirchhoff ,  Srikanth Ronanki

IPC: G06F40/166 ,  G06F21/62 ,  G06F40/279 ,  G10L15/16 ,  G10L15/22

Abstract: Portions of text data generated from inverse text normalization may be redacted. Text data for redaction may be obtained. One or more inverse text normalization models may be applied to the text data to generate normalized text data. A machine learning model, trained to recognize text for redaction, may be applied to identify portions of the normalized text data for redaction. The identified portions may be redacted and the redacted normalized text provided to a destination.