中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

13 records (took 0.013 seconds)

    Phrase extraction for ASR models
    2.
    发明授权
    Phrase extraction for ASR models 有权

    公开(公告)号：US12260875B2

    公开(公告)日：2025-03-25

    申请号：US18609362

    申请日：2024-03-19

    Applicant: Google LLC

    Inventor： Ehsan Amid ,  Om Dipakbhai Thakkar ,  Rajiv Mathews ,  Francoise Beaufays

    IPC: G10L21/0332 ,  G10L15/06 ,  G10L15/08 ,  G10L21/10

    Abstract: A method of phrase extraction for ASR models includes obtaining audio data characterizing an utterance and a corresponding ground-truth transcription of the utterance and modifying the audio data to obfuscate a particular phrase recited in the utterance. The method also includes processing, using a trained ASR model, the modified audio data to generate a predicted transcription of the utterance, and determining whether the predicted transcription includes the particular phrase by comparing the predicted transcription of the utterance to the ground-truth transcription of the utterance. When the predicted transcription includes the particular phrase, the method includes generating an output indicating that the trained ASR model leaked the particular phrase from a training data set used to train the ASR model.

    Phrase extraction for ASR models
    3.
    发明授权
    Phrase extraction for ASR models 有权

    公开(公告)号：US11955134B2

    公开(公告)日：2024-04-09

    申请号：US17643848

    申请日：2021-12-13

    Applicant: Google LLC

    Inventor： Ehsan Amid ,  Om Thakkar ,  Rajiv Mathews ,  Francoise Beaufays

    IPC: G10L21/0332 ,  G10L15/06 ,  G10L15/08 ,  G10L21/10

    CPC classification number: G10L21/0332 ,  G10L15/063 ,  G10L15/08 ,  G10L21/10

    Abstract: A method of phrase extraction for ASR models includes obtaining audio data characterizing an utterance and a corresponding ground-truth transcription of the utterance and modifying the audio data to obfuscate a particular phrase recited in the utterance. The method also includes processing, using a trained ASR model, the modified audio data to generate a predicted transcription of the utterance, and determining whether the predicted transcription includes the particular phrase by comparing the predicted transcription of the utterance to the ground-truth transcription of the utterance. When the predicted transcription includes the particular phrase, the method includes generating an output indicating that the trained ASR model leaked the particular phrase from a training data set used to train the ASR model.

    Unified Sample Reweighting Framework for Learning with Noisy Data and for Learning Difficult Examples or Groups
    4.
    发明申请
    Unified Sample Reweighting Framework for Learning with Noisy Data and for Learning Difficult Examples or Groups 有权

    公开(公告)号：US20230044078A1

    公开(公告)日：2023-02-09

    申请号：US17816197

    申请日：2022-07-29

    Applicant: Google LLC

    Inventor： Abhishek Kumar ,  Ehsan Amid

    IPC: G06N20/00

    Abstract: A method includes receiving training data for a machine learning model, the training data comprising a plurality of training examples and a corresponding plurality of labels. The method further includes dividing the training data into a plurality of training batches. For each training batch of the plurality of training batches, the method additionally includes learning a weight for each training example in the training batch that minimizes a sum of weighted losses for the training batch subject to a divergence constraint, where the divergence constraint limits a divergence of the learned weights for the training batch from a reference distribution, where the divergence is determined according to a chosen divergence measure. The method also includes training the machine learning model with each training batch of the plurality of training batches using the learned weight for each training example in the training batch. The method additionally includes providing the trained machine learning model.

    Knowledge Distillation Via Learning to Predict Principal Components Coefficients
    5.
    发明申请
    Knowledge Distillation Via Learning to Predict Principal Components Coefficients 有权

    公开(公告)号：US20250005453A1

    公开(公告)日：2025-01-02

    申请号：US18710814

    申请日：2022-12-12

    Applicant: Google LLC

    Inventor： Ehsan Amid ,  Christopher James Fifty ,  Manfred Klaus Warmuth ,  Rohan Anil

    IPC: G06N20/00

    Abstract: Provided is an approach for knowledge distillation based on exporting Principal Components approximations (e.g., Bregman representations) of one or more layer-wise representations of the teacher model. In particular, the present disclosure provides an extension to the original Bregman PCA formulation by incorporating a mean vector and orthonormalizing the principal directions with respect to the geometry of the local convex function around the mean. This extended formulation allows viewing the learned representation as a dense layer, thus casting the problem as learning the linear coefficients of the compressed examples, as the input to this layer, by the student network. Example empirical data indicates that example implementations of the approach improve performance when compared to typical teacher-student training using soft labels.

    Heterogeneous Federated Learning Via Multi-Directional Knowledge Distillation
    9.
    发明公开
    Heterogeneous Federated Learning Via Multi-Directional Knowledge Distillation 审中-公开

    公开(公告)号：US20240249193A1

    公开(公告)日：2024-07-25

    申请号：US18417947

    申请日：2024-01-19

    Applicant: Google LLC

    Inventor： Jared Alexander Lichtarge ,  Rajiv Mathews ,  Rohan Anil ,  Ehsan Amid ,  Shankar Kumar

    IPC: G06N20/00

    CPC classification number: G06N20/00

    Abstract: Generally, the present disclosure is directed to enhanced federated learning (FL) that employs a set of clients with varying amounts of computational resources (e.g., system memory, storage, and processing bandwidth). To overcome limitations of conventional FL methods that employ a set of clients with varying amounts of computational resources, the embodiments run multi-directional knowledge distillation between the server models produced by each federated averaging (FedAvg) pool, using unlabeled server data as the distillation dataset. By co-distilling the two (or more) models frequently over the course of FedAvg rounds, information is shared between the pools without sharing model parameters. This leads to increased performance and faster convergence (in fewer federated rounds).

    Phrase Extraction for ASR Models
    10.
    发明公开
    Phrase Extraction for ASR Models 审中-公开

    公开(公告)号：US20240221772A1

    公开(公告)日：2024-07-04

    申请号：US18609362

    申请日：2024-03-19

    Applicant: Google LLC

    Inventor： Ehsan Amid ,  Om Dipakbhai Thakkar ,  Rajiv Mathews ,  Francoise Beaufays

    IPC: G10L21/0332 ,  G10L15/06 ,  G10L15/08 ,  G10L21/10

    CPC classification number: G10L21/0332 ,  G10L15/063 ,  G10L15/08 ,  G10L21/10

    Abstract: A method of phrase extraction for ASR models includes obtaining audio data characterizing an utterance and a corresponding ground-truth transcription of the utterance and modifying the audio data to obfuscate a particular phrase recited in the utterance. The method also includes processing, using a trained ASR model, the modified audio data to generate a predicted transcription of the utterance, and determining whether the predicted transcription includes the particular phrase by comparing the predicted transcription of the utterance to the ground-truth transcription of the utterance. When the predicted transcription includes the particular phrase, the method includes generating an output indicating that the trained ASR model leaked the particular phrase from a training data set used to train the ASR model.

Patent Agency Ranking