-
公开(公告)号：US20240203406A1
公开(公告)日：2024-06-20
申请号：US18065685
申请日：2022-12-14
申请人： Google LLC
发明人： Soheil Khorram , Anshuman Tripathi , Kim Jaeyoung , Han Lu , Qian Zhang , Hasim Sak
IPC分类号： G10L15/183 , G10L15/06 , G10L15/22
CPC分类号： G10L15/183 , G10L15/063 , G10L15/22
摘要： A method includes receiving a sequence of acoustic frames extracted from unlabeled audio samples that correspond to spoken utterances not paired with any corresponding transcriptions. The method also includes generating, using a supervised audio encoder, a target higher order feature representation for a corresponding acoustic frame. The method also includes augmenting the sequence of acoustic frames and generating, as output form an unsupervised audio encoder, a predicted higher order feature representation for a corresponding augmented acoustic frame in the sequence of augmented acoustic frames. The method also includes determining an unsupervised loss term based on the target higher order feature representation and the predicted higher order feature representation and updating parameters of the speech recognition model based on the unsupervised loss term.
-
公开(公告)号：US20240290322A1
公开(公告)日：2024-08-29
申请号：US18587860
申请日：2024-02-26
申请人： Google LLC
发明人： JAEYOUNG Kim , Han Lu , Soheil Khorram , Anshuman Tripathi , Qian Zhang , Hasim Sak
IPC分类号： G10L15/06
CPC分类号： G10L15/063
摘要： A method of training an accent recognition model includes receiving a corpus of training utterances spoken across various accents, each training utterance in the corpus including training audio features characterizing the training utterance, and executing a training process to train the accent recognition model on the corpus of training utterances to teach the accent recognition model to learn how to predict accent representations from the training audio features. The accent recognition model includes one or more strided convolution layers, a stack of multi-headed attention layers, and a pooling layer configured to generate a corresponding accent representation.
-
3.发明授权公开(公告)号：US11741947B2
公开(公告)日：2023-08-29
申请号：US17210465
申请日：2021-03-23
申请人： Google LLC
发明人： Anshuman Tripathi , Hasim Sak , Han Lu , Qian Zhang , Jaeyoung Kim
CPC分类号： G10L15/16 , G06N3/04 , G06N3/088 , G10L15/063 , G10L15/197 , G10L15/22 , G10L15/30
摘要： A transformer-transducer model for unifying streaming and non-streaming speech recognition includes an audio encoder, a label encoder, and a joint network. The audio encoder receives a sequence of acoustic frames, and generates, at each of a plurality of time steps, a higher order feature representation for a corresponding acoustic frame. The label encoder receives a sequence of non-blank symbols output by a final softmax layer, and generates, at each of the plurality of time steps, a dense representation. The joint network receives the higher order feature representation and the dense representation at each of the plurality of time steps, and generates a probability distribution over possible speech recognition hypothesis. The audio encoder of the model further includes a neural network having an initial stack of transformer layers trained with zero look ahead audio context, and a final stack of transformer layers trained with a variable look ahead audio context.
-
公开(公告)号：US20220310097A1
公开(公告)日：2022-09-29
申请号：US17644377
申请日：2021-12-15
申请人： Google LLC
发明人： Jaeyoung Kim , Han Lu , Anshuman Tripathi , Qian Zhang , Hasim Sak
摘要： A streaming speech recognition model includes an audio encoder configured to receive a sequence of acoustic frames and generate a higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The streaming speech recognition model also includes a label encoder configured to receive a sequence of non-blank symbols output by a final softmax layer and generate a dense representation. The streaming speech recognition model also includes a joint network configured to receive the higher order feature representation generated by the audio encoder and the dense representation generated by the label encoder and generate a probability distribution over possible speech recognition hypotheses. Here, the streaming speech recognition model is trained using self-alignment to reduce prediction delay by encouraging an alignment path that is one frame left from a reference forced-alignment frame.
-
公开(公告)号：US20210343273A1
公开(公告)日：2021-11-04
申请号：US16865075
申请日：2020-05-01
申请人： Google LLC
发明人： Anshuman Tripathi , Han Lu , Hasim Sak
摘要： A method for training a speech recognition model with a loss function includes receiving an audio signal including a first segment corresponding to audio spoken by a first speaker, a second segment corresponding to audio spoken by a second speaker, and an overlapping region where the first segment overlaps the second segment. The overlapping region includes a known start time and a known end time. The method also includes generating a respective masked audio embedding for each of the first and second speakers. The method also includes applying a masking loss after the known end time to the respective masked audio embedding for the first speaker when the first speaker was speaking prior to the known start time, or applying the masking loss prior to the known start time when the first speaker was speaking after the known end time.
-
公开(公告)号：US20240135934A1
公开(公告)日：2024-04-25
申请号：US18483492
申请日：2023-10-09
申请人： Google LLC
发明人： Guanlong Zhao , Quan Wang , Han Lu , Yiling Huang , Jason Pelecanos
摘要： A method includes obtaining a multi-utterance training sample that includes audio data characterizing utterances spoken by two or more different speakers and obtaining ground-truth speaker change intervals indicating time intervals in the audio data where speaker changes among the two or more different speakers occur. The method also includes processing the audio data to generate a sequence of predicted speaker change tokens using a sequence transduction model. For each corresponding predicted speaker change token, the method includes labeling the corresponding predicted speaker change token as correct when the predicted speaker change token overlaps with one of the ground-truth speaker change intervals. The method also includes determining a precision metric of the sequence transduction model based on a number of the predicted speaker change tokens labeled as correct and a total number of the predicted speaker change tokens in the sequence of predicted speaker change tokens.
-
公开(公告)号：US11961515B2
公开(公告)日：2024-04-16
申请号：US17644337
申请日：2021-12-14
申请人： Google LLC
发明人： Jaeyoung Kim , Soheil Khorram , Hasim Sak , Anshuman Tripathi , Han Lu , Qian Zhang
CPC分类号： G10L15/16 , G06N3/088 , G10L15/1815
摘要： A method includes receiving a plurality of unlabeled audio samples corresponding to spoken utterances not paired with corresponding transcriptions. At a target branch of a contrastive Siamese network, the method also includes generating a sequence of encoder outputs for the plurality of unlabeled audio samples and modifying time characteristics of the encoder outputs to generate a sequence of target branch outputs. At an augmentation branch of a contrastive Siamese network, the method also includes performing augmentation on the unlabeled audio samples, generating a sequence of augmented encoder outputs for the augmented unlabeled audio samples, and generating predictions of the sequence of target branch outputs generated at the target branch. The method also includes determining an unsupervised loss term based on target branch outputs and predictions of the sequence of target branch outputs. The method also includes updating parameters of the audio encoder based on the unsupervised loss term.
-
公开(公告)号：US20230368779A1
公开(公告)日：2023-11-16
申请号：US18357225
申请日：2023-07-24
申请人： Google LLC
发明人： Anshuman Tripathi , Hasim Sak , Han Lu , Qian Zhang , Jaeyoung Kim
CPC分类号： G10L15/16 , G06N3/088 , G10L15/063 , G10L15/22 , G10L15/30 , G06N3/04 , G10L15/197
摘要： A transformer-transducer model for unifying streaming and non-streaming speech recognition includes an audio encoder, a label encoder, and a joint network. The audio encoder receives a sequence of acoustic frames, and generates, at each of a plurality of time steps, a higher order feature representation for a corresponding acoustic frame. The label encoder receives a sequence of non-blank symbols output by a final softmax layer, and generates, at each of the plurality of time steps, a dense representation. The joint network receives the higher order feature representation and the dense representation at each of the plurality of time steps, and generates a probability distribution over possible speech recognition hypothesis. The audio encoder of the model further includes a neural network having an initial stack of transformer layers trained with zero look ahead audio context, and a final stack of transformer layers trained with a variable look ahead audio context.
-
公开(公告)号：US11521595B2
公开(公告)日：2022-12-06
申请号：US16865075
申请日：2020-05-01
申请人： Google LLC
发明人： Anshuman Tripathi , Han Lu , Hasim Sak
摘要： A method for training a speech recognition model with a loss function includes receiving an audio signal including a first segment corresponding to audio spoken by a first speaker, a second segment corresponding to audio spoken by a second speaker, and an overlapping region where the first segment overlaps the second segment. The overlapping region includes a known start time and a known end time. The method also includes generating a respective masked audio embedding for each of the first and second speakers. The method also includes applying a masking loss after the known end time to the respective masked audio embedding for the first speaker when the first speaker was speaking prior to the known start time, or applying the masking loss prior to the known start time when the first speaker was speaking after the known end time.
-
10.发明申请公开(公告)号：US20220108689A1
公开(公告)日：2022-04-07
申请号：US17210465
申请日：2021-03-23
申请人： Google LLC
发明人： Anshuman Tripathi , Hasim Sak , Han Lu , Qian Zhang , Jaeyoung Kim
摘要： A transformer-transducer model for unifying streaming and non-streaming speech recognition includes an audio encoder, a label encoder, and a joint network. The audio encoder receives a sequence of acoustic frames, and generates, at each of a plurality of time steps, a higher order feature representation for a corresponding acoustic frame. The label encoder receives a sequence of non-blank symbols output by a final softmax layer, and generates, at each of the plurality of time steps, a dense representation. The joint network receives the higher order feature representation and the dense representation at each of the plurality of time steps, and generates a probability distribution over possible speech recognition hypothesis. The audio encoder of the model further includes a neural network having an initial stack of transformer layers trained with zero look ahead audio context, and a final stack of transformer layers trained with a variable look ahead audio context.
-
-
-
-
-
-
-
-
-
搜索结果
15 条记录 (耗时 0.022 秒)
