公开(公告)号：US10699695B1

公开(公告)日：2020-06-30

申请号：US16023370

申请日：2018-06-29

Applicant: Amazon Technologies, Inc.

Inventor： Adam Franciszek Nadolski ,  Daniel Korzekwa ,  Thomas Edward Merritt ,  Marco Nicolis ,  Bartosz Putrycz ,  Roberto Barra Chicote ,  Rafal Kuklinski ,  Wiktor Dolecki

IPC: G10L13/10 ,  G10L13/06 ,  G10L13/047

Abstract: During text-to-speech processing, audio data corresponding to a word part, word, or group of words is generated using a trained model and used by a unit selection engine to create output audio. The audio data is generated at least when an input word is unrecognized or when a cost of a unit selection is too high.

公开(公告)号：US10692484B1

公开(公告)日：2020-06-23

申请号：US16007757

申请日：2018-06-13

Applicant: Amazon Technologies, Inc.

Inventor： Thomas Edward Merritt ,  Adam Franciszek Nadolski ,  Nishant Prateek ,  Bartosz Putrycz ,  Roberto Barra Chicote ,  Vatsal Aggarwal ,  Andrew Paul Breen

IPC: G10L13/04 ,  G10L13/08 ,  G10L25/24 ,  G10L25/60 ,  G10L13/047

Abstract: A speech model is trained using multi-task learning. A first task may correspond to how well predicted audio matches training audio; a second task may correspond to a metric of perceived audio quality. The speech model may include, during training, layers related to the second task that are discarded at runtime.

公开(公告)号：US20250104693A1

公开(公告)日：2025-03-27

申请号：US18474484

申请日：2023-09-26

Applicant: Amazon Technologies, Inc.

Inventor： Constantinos Papayiannis ,  Roberto Barra Chicote ,  Trevor Michael Wood ,  James Garnet Droppo

IPC: G10L13/10 ,  G10L13/047 ,  G10L25/18

Abstract: Techniques for using a language model (e.g., a large language model (LLM)) to generate a natural language response to a user input and prosody information (e.g., voice characteristics associated with a synthetic voice to output the natural language response to the user) are described. The prosody information may correspond to a natural language (e.g., text or tokenized) description, a spectrogram, and/or a latent representation of the voice characteristic(s) associated with the natural language response. In some embodiments, the natural language response and the prosody information may be generated by different portions of layers of the language model. In such embodiments, the output of the layer(s) of the language model configured to generate the natural language response may be provided to the layer(s) of the language model configured to generate the prosody information and the output may be used to generate the prosody information, and vice versa.

公开(公告)号：US12100383B1

公开(公告)日：2024-09-24

申请号：US17707203

申请日：2022-03-29

Applicant: Amazon Technologies, Inc.

Inventor： Abdelhamid Ezzerg ,  Piotr Tadeusz Bilinski ,  Thomas Edward Merritt ,  Roberto Barra Chicote ,  Daniel Korzekwa ,  Kamil Pokora

IPC: G10L13/047 ,  G06N3/045 ,  G10L25/30

CPC classification number: G10L13/047 ,  G06N3/045 ,  G10L25/30

Abstract: Voice customization is an application of voice synthesis that involves synthesizing speech having certain voice characteristics, and/or modifying the voice characteristics of human speech. Certain techniques for voice customization may be used in conjunction with compressing speech for storage and/or transmission. For example, speech may be received at a first device and transformed into a latent representation and/or compressed for storage and/or transmission to a second device. The system may use normalizing flows to transform the source audio to a latent representation having a desired variable distribution, and to transform the latent representation back into audio data. A flow model may conditioned using first speech attributes when transforming the source audio, and an inverse flow model may use second speech attributes when transforming the latent representation back into audio data. The first and/or second speech attributes may be modified to alter voice characteristics of the transmitted speech.

公开(公告)号：US11017763B1

公开(公告)日：2021-05-25

申请号：US16712466

申请日：2019-12-12

Applicant: Amazon Technologies, Inc.

Inventor： Vatsal Aggarwal ,  Nishant Prateek ,  Roberto Barra Chicote ,  Andrew Paul Breen

IPC: G10L15/22 ,  G10L15/26 ,  G10L13/08 ,  G10L13/047 ,  G10L13/033

Abstract: During text-to-speech processing, a sequence-to-sequence neural network model may process text data and determine corresponding spectrogram data. A normalizing flow component may then process this spectrogram data to predict corresponding phase data. An inverse Fourier transform may then be performed on the spectrogram and phase data to create an audio waveform that includes speech corresponding to the text.

公开(公告)号：US10706837B1

公开(公告)日：2020-07-07

申请号：US16007811

申请日：2018-06-13

Applicant: Amazon Technologies, Inc.

Inventor： Roberto Barra Chicote ,  Adam Franciszek Nadolski ,  Thomas Edward Merritt ,  Bartosz Putrycz ,  Andrew Paul Breen

IPC: G10L13/033 ,  G10L13/04 ,  G10L13/10

Abstract: A speech model includes a sub-model corresponding to a vocal attribute. The speech model generates an output waveform using a sample model, which receives text data, and a conditioning model, which receives text metadata and produces a prosody output for use by the sample model. If, during training or runtime, a different vocal attribute is desired or needed, the sub-model is re-trained or switched to a different sub-model corresponding to the different vocal attribute.

公开(公告)号：US10475438B1

公开(公告)日：2019-11-12

申请号：US15447919

申请日：2017-03-02

Applicant: Amazon Technologies, Inc.

Inventor： Roberto Barra Chicote ,  Javier Latorre ,  Adam Franciszek Nadolski ,  Viacheslav Klimkov ,  Thomas Edward Merritt

IPC: G10L13/10 ,  G10L13/033 ,  G10L13/047

Abstract: A text-to-speech (TTS) system that is capable of considering characteristics of various portions of text data in order to create continuity between segments of synthesized speech. The system can analyze text portions of a work and create feature vectors including data corresponding to characteristics of the individual portions and/or the overall work. A TTS processing component can then consider feature vector(s) from other portions when performing TTS processing on text of a first portion, thus giving the TTS component some intelligence regarding other portions of the work, which can then result in more continuity between synthesized speech segments.