利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

42 条记录 (耗时 0.025 秒)

    EPHEMERAL LEARNING OF MACHINE LEARNING MODEL(S)
    11.
    发明公开
    EPHEMERAL LEARNING OF MACHINE LEARNING MODEL(S) 审中-公开

    公开(公告)号：US20230156248A1

    公开(公告)日：2023-05-18

    申请号：US17533779

    申请日：2021-11-23

    申请人： GOOGLE LLC

    发明人： Françoise Beaufays ,  Khe Chai Sim ,  Trevor Strohman ,  Oren Litvin

    IPC分类号： H04N21/233 ,  G06N20/00 ,  G06K9/62 ,  H04N21/232

    CPC分类号： H04N21/233 ,  G06N20/00 ,  G06K9/6256 ,  H04N21/232

    摘要： Implementations disclosed herein are directed to ephemeral learning of machine learning (“ML”) model(s) based on gradient(s) generated at a remote system (e.g., remote server(s)). Processor(s) of the remote system can receive stream(s) of audio data capturing spoken utterance(s) from a client device of a user. A fulfillment pipeline can process the stream(s) of audio data to cause certain fulfillment(s) of the spoken utterance(s) to be performed. Meanwhile, a training pipeline can process the stream(s) of audio data to generate gradient(s) using unsupervised learning techniques. Subsequent to the processing by the fulfillment pipeline and/or the training pipeline, the stream(s) of audio data are discarded by the remote system. Accordingly, the ML model(s) can be trained at the remote system without storing or logging of the stream(s) of audio data by non-transient memory thereof, thereby providing more efficient training mechanisms for training the ML model(s) and also increasing security of user data.

    Emitting word timings with end-to-end models
    14.
    发明授权
    Emitting word timings with end-to-end models 有权

    公开(公告)号：US12027154B2

    公开(公告)日：2024-07-02

    申请号：US18167050

    申请日：2023-02-09

    申请人： Google LLC

    发明人： Tara N. Sainath ,  Basilio Garcia Castillo ,  David Rybach ,  Trevor Strohman ,  Ruoming Pang

    IPC分类号： G10L25/30 ,  G10L15/06 ,  G10L25/78

    CPC分类号： G10L15/063 ,  G10L25/30 ,  G10L25/78

    摘要： A method includes receiving a training example that includes audio data representing a spoken utterance and a ground truth transcription. For each word in the spoken utterance, the method also includes inserting a placeholder symbol before the respective word identifying a respective ground truth alignment for a beginning and an end of the respective word, determining a beginning word piece and an ending word piece, and generating a first constrained alignment for the beginning word piece and a second constrained alignment for the ending word piece. The first constrained alignment is aligned with the ground truth alignment for the beginning of the respective word and the second constrained alignment is aligned with the ground truth alignment for the ending of the respective word. The method also includes constraining an attention head of a second pass decoder by applying the first and second constrained alignments.

    EFFICIENT STREAMING NON-RECURRENT ON-DEVICE END-TO-END MODEL
    16.
    发明公开
    EFFICIENT STREAMING NON-RECURRENT ON-DEVICE END-TO-END MODEL 审中-公开

    公开(公告)号：US20230343328A1

    公开(公告)日：2023-10-26

    申请号：US18336211

    申请日：2023-06-16

    申请人： Google LLC

    发明人： Tara Sainath ,  Arun Narayanan ,  Rami Botros ,  Yanzhang He ,  Ehsan Variani ,  Cyril Allauzen ,  David Rybach ,  Ruoming Pang ,  Trevor Strohman

    IPC分类号： G10L15/06 ,  G10L15/02 ,  G10L15/22 ,  G10L15/30

    CPC分类号： G10L15/063 ,  G10L15/02 ,  G10L15/22 ,  G10L15/30

    摘要： An ASR model includes a first encoder configured to receive a sequence of acoustic frames and generate a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The ASR model also includes a second encoder configured to receive the first higher order feature representation generated by the first encoder at each of the plurality of output steps and generate a second higher order feature representation for a corresponding first higher order feature frame. The ASR model also includes a decoder configured to receive the second higher order feature representation generated by the second encoder at each of the plurality of output steps and generate a first probability distribution over possible speech recognition hypothesis. The ASR model also includes a language model configured to receive the first probability distribution over possible speech hypothesis and generate a rescored probability distribution.

    Deliberation by Text-Only and Semi-Supervised Training
    18.
    发明公开
    Deliberation by Text-Only and Semi-Supervised Training 审中-公开

    公开(公告)号：US20230298563A1

    公开(公告)日：2023-09-21

    申请号：US18186157

    申请日：2023-03-18

    申请人： Google LLC

    发明人： Ke Hu ,  Tara N. Sainath ,  Yanzhang He ,  Rohit Prabhavalkar ,  Sepand Mavandadi ,  Weiran Wang ,  Trevor Strohman

    IPC分类号： G10L13/08 ,  G10L15/16 ,  G10L15/06

    CPC分类号： G10L13/08 ,  G10L15/16 ,  G10L15/063

    摘要： A method of text-only and semi-supervised training for deliberation includes receiving training data including unspoken textual utterances that are each not paired with any corresponding spoken utterance of non-synthetic speech, and training a deliberation model that includes a text encoder and a deliberation decoder on the unspoken textual utterances. The method also includes receiving, at the trained deliberation model, first-pass hypotheses and non-causal acoustic embeddings. The first-pass hypotheses is generated by a recurrent neural network-transducer (RNN-T) decoder for the non-causal acoustic embeddings encoded by a non-causal encoder. The method also includes encoding, using the text encoder, the first-pass hypotheses generated by the RNN-T decoder, and generating, using the deliberation decoder attending to both the first-pass hypotheses and the non-causal acoustic embeddings, second-pass hypotheses.

    METHODS AND SYSTEMS FOR REDUCING LATENCY IN AUTOMATED ASSISTANT INTERACTIONS
    19.
    发明申请
    METHODS AND SYSTEMS FOR REDUCING LATENCY IN AUTOMATED ASSISTANT INTERACTIONS 有权

    公开(公告)号：US20220351720A1

    公开(公告)日：2022-11-03

    申请号：US17243232

    申请日：2021-04-28

    申请人： Google LLC

    发明人： Lior Alon ,  Rafael Goldfarb ,  Dekel Auster ,  Dan Rasin ,  Michael Andrew Goodman ,  Trevor Strohman ,  Nino Tasca ,  Valerie Nygaard ,  Jaclyn Konzelmann

    IPC分类号： G10L15/22 ,  G10L15/28 ,  G10L15/18 ,  G10L15/08 ,  G06F3/16

    摘要： Implementations described herein relate to reducing latency in automated assistant interactions. In some implementations, a client device can receive audio data that captures a spoken utterance of a user. The audio data can be processed to determine an assistant command to be performed by an automated assistant. The assistant command can be processed, using a latency prediction model, to generate a predicted latency to fulfill the assistant command. Further, the client device (or the automated assistant) can determine, based on the predicted latency, whether to audibly render pre-cached content for presentation to the user prior to audibly rendering content that is responsive to the spoken utterance. The pre-cached content can be tailored to the assistant command and audibly rendered for presentation to the user while the content is being obtained, and the content can be audibly rendered for presentation to the user subsequent to the pre-cached content.

    Cascaded Encoders for Simplified Streaming and Non-Streaming ASR
    20.
    发明申请
    Cascaded Encoders for Simplified Streaming and Non-Streaming ASR 有权

    公开(公告)号：US20220122622A1

    公开(公告)日：2022-04-21

    申请号：US17237021

    申请日：2021-04-21

    申请人： Google LLC

    发明人： Arun Narayanan ,  Tara Sainath ,  Chung-Cheng Chiu ,  Ruoming Pang ,  Rohit Prabhavalkar ,  Jiahui Yu ,  Ehsan Variani ,  Trevor Strohman

    IPC分类号： G10L19/16 ,  G10L15/00 ,  G10L25/30 ,  G06N3/08

    摘要： An automated speech recognition (ASR) model includes a first encoder, a second encoder, and a decoder. The first encoder receives, as input, a sequence of acoustic frames, and generates, at each of a plurality of output steps, a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The second encoder receives, as input, the first higher order feature representation generated by the first encoder at each of the plurality of output steps, and generates, at each of the plurality of output steps, a second higher order feature representation for a corresponding first higher order feature frame. The decoder receives, as input, the second higher order feature representation generated by the second encoder at each of the plurality of output steps, and generates, at each of the plurality of time steps, a first probability distribution over possible speech recognition hypotheses.