专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US11574639B2 Hypothesis stitcher for speech recognition of long-form audio 有权
公开(公告)号：US11574639B2
公开(公告)日：2023-02-07
申请号：US17127938
申请日：2020-12-18
申请人： Microsoft Technology Licensing, LLC
发明人： Naoyuki Kanda , Xuankai Chang , Yashesh Gaur , Xiaofei Wang , Zhong Meng , Takuya Yoshioka
IPC分类号： G10L15/00 , G10L17/02 , G10L15/22 , G10L15/26 , G10L19/022 , G10L21/0272
摘要： A hypothesis stitcher for speech recognition of long-form audio provides superior performance, such as higher accuracy and reduced computational cost. An example disclosed operation includes: segmenting the audio stream into a plurality of audio segments; identifying a plurality of speakers within each of the plurality of audio segments; performing automatic speech recognition (ASR) on each of the plurality of audio segments to generate a plurality of short-segment hypotheses; merging at least a portion of the short-segment hypotheses into a first merged hypothesis set; inserting stitching symbols into the first merged hypothesis set, the stitching symbols including a window change (WC) symbol; and consolidating, with a network-based hypothesis stitcher, the first merged hypothesis set into a first consolidated hypothesis. Multiple variations are disclosed, including alignment-based stitchers and serialized stitchers, which may operate as speaker-specific stitchers or multi-speaker stitchers, and may further support multiple options for differing hypothesis configurations.

2. 发明授权

US11935542B2 Hypothesis stitcher for speech recognition of long-form audio 有权
公开(公告)号：US11935542B2
公开(公告)日：2024-03-19
申请号：US18157070
申请日：2023-01-19
申请人： Microsoft Technology Licensing, LLC
发明人： Naoyuki Kanda , Xuankai Chang , Yashesh Gaur , Xiaofei Wang , Zhong Meng , Takuya Yoshioka
IPC分类号： G10L15/00 , G10L15/22 , G10L15/26 , G10L17/02 , G10L19/022 , G10L21/0272
CPC分类号： G10L17/02 , G10L15/22 , G10L15/26 , G10L19/022 , G10L21/0272
摘要： A hypothesis stitcher for speech recognition of long-form audio provides superior performance, such as higher accuracy and reduced computational cost. An example disclosed operation includes: segmenting the audio stream into a plurality of audio segments; identifying a plurality of speakers within each of the plurality of audio segments; performing automatic speech recognition (ASR) on each of the plurality of audio segments to generate a plurality of short-segment hypotheses; merging at least a portion of the short-segment hypotheses into a first merged hypothesis set; inserting stitching symbols into the first merged hypothesis set, the stitching symbols including a window change (WC) symbol; and consolidating, with a network-based hypothesis stitcher, the first merged hypothesis set into a first consolidated hypothesis. Multiple variations are disclosed, including alignment-based stitchers and serialized stitchers, which may operate as speaker-specific stitchers or multi-speaker stitchers, and may further support multiple options for differing hypothesis configurations.

3. 发明授权

US11107460B2 Adversarial speaker adaptation 有权
公开(公告)号：US11107460B2
公开(公告)日：2021-08-31
申请号：US16460027
申请日：2019-07-02
申请人： Microsoft Technology Licensing, LLC
发明人： Zhong Meng , Jinyu Li , Yifan Gong
IPC分类号： G10L15/06 , G10L15/02 , G10L15/22 , G10L15/16
摘要： Embodiments are associated with a speaker-independent acoustic model capable of classifying senones based on input speech frames and on first parameters of the speaker-independent acoustic model, a speaker-dependent acoustic model capable of classifying senones based on input speech frames and on second parameters of the speaker-dependent acoustic model, and a discriminator capable of receiving data from the speaker-dependent acoustic model and data from the speaker-independent acoustic model and outputting a prediction of whether received data was generated by the speaker-dependent acoustic model based on third parameters. The second parameters are initialized based on the first parameters, the second parameters are trained based on input frames of a target speaker to minimize a senone classification loss associated with the second parameters, a portion of the second parameters are trained based on the input frames of the target speaker to maximize a discrimination loss associated with the discriminator, and the third parameters are trained based on the input frames of the target speaker to minimize the discrimination loss.

4. 发明授权

US11915686B2 Speaker adaptation for attention-based encoder-decoder 有权
公开(公告)号：US11915686B2
公开(公告)日：2024-02-27
申请号：US17568875
申请日：2022-01-05
申请人： Microsoft Technology Licensing, LLC
发明人： Zhong Meng , Yashesh Gaur , Jinyu Li , Yifan Gong
IPC分类号： G10L15/065 , G10L15/06 , G10L15/22 , G10L19/00
CPC分类号： G10L15/065 , G10L15/063 , G10L15/22 , G10L19/00
摘要： Embodiments are associated with a speaker-independent attention-based encoder-decoder model to classify output tokens based on input speech frames, the speaker-independent attention-based encoder-decoder model associated with a first output distribution, and a speaker-dependent attention-based encoder-decoder model to classify output tokens based on input speech frames, the speaker-dependent attention-based encoder-decoder model associated with a second output distribution. The second attention-based encoder-decoder model is trained to classify output tokens based on input speech frames of a target speaker and simultaneously trained to maintain a similarity between the first output distribution and the second output distribution.

5. 发明授权

US11586930B2 Conditional teacher-student learning for model training 有权
公开(公告)号：US11586930B2
公开(公告)日：2023-02-21
申请号：US16410741
申请日：2019-05-13
申请人： Microsoft Technology Licensing, LLC
发明人： Zhong Meng , Jinyu Li , Yong Zhao , Yifan Gong
IPC分类号： G06N3/08 , G10L15/183 , G06N3/04 , G10L15/16 , G06N3/084
摘要： Embodiments are associated with conditional teacher-student model training. A trained teacher model configured to perform a task may be accessed and an untrained student model may be created. A model training platform may provide training data labeled with ground truths to the teacher model to produce teacher posteriors representing the training data. When it is determined that a teacher posterior matches the associated ground truth label, the platform may conditionally use the teacher posterior to train the student model. When it is determined that a teacher posterior does not match the associated ground truth label, the platform may conditionally use the ground truth label to train the student model. The models might be associated with, for example, automatic speech recognition (e.g., in connection with domain adaptation and/or speaker adaptation).

6. 发明授权

US11232782B2 Speaker adaptation for attention-based encoder-decoder 有权
公开(公告)号：US11232782B2
公开(公告)日：2022-01-25
申请号：US16675515
申请日：2019-11-06
申请人： Microsoft Technology Licensing, LLC
发明人： Zhong Meng , Yashesh Gaur , Jinyu Li , Yifan Gong
IPC分类号： G10L15/065 , G10L15/06 , G10L19/00 , G10L15/22
摘要： Embodiments are associated with a speaker-independent attention-based encoder-decoder model to classify output tokens based on input speech frames, the speaker-independent attention-based encoder-decoder model associated with a first output distribution, a speaker-dependent attention-based encoder-decoder model to classify output tokens based on input speech frames, the speaker-dependent attention-based encoder-decoder model associated with a second output distribution, training of the second attention-based encoder-decoder model to classify output tokens based on input speech frames of a target speaker and simultaneously training the speaker-dependent attention-based encoder-decoder model to maintain a similarity between the first output distribution and the second output distribution, and performing automatic speech recognition on speech frames of the target speaker using the trained speaker-dependent attention-based encoder-decoder model.

7. 发明授权

US10347241B1 Speaker-invariant training via adversarial learning 有权
公开(公告)号：US10347241B1
公开(公告)日：2019-07-09
申请号：US15934566
申请日：2018-03-23
申请人： Microsoft Technology Licensing, LLC
发明人： Zhong Meng , Vadim Aleksandrovich Mazalov , Yifan Gong , Yong Zhao , Zhuo Chen , Jinyu Li
IPC分类号： G10L25/30 , G10L15/02 , G10L15/16 , G10L15/06
摘要： Systems and methods can be implemented to conduct speaker-invariant training for speech recognition in a variety of applications. An adversarial multi-task learning scheme for speaker-invariant training can be implemented, aiming at actively curtailing the inter-talker feature variability, while maximizing its senone discriminability to enhance the performance of a deep neural network (DNN) based automatic speech recognition system. In speaker-invariant training, a DNN acoustic model and a speaker classifier network can be jointly optimized to minimize the senone (triphone state) classification loss, and simultaneously mini-maximize the speaker classification loss. A speaker invariant and senone-discriminative intermediate feature is learned through this adversarial multi-task learning, which can be applied to an automatic speech recognition system. Additional systems and methods are disclosed.

8. 发明授权

US11984127B2 Training and using a transcript generation model on a multi-speaker audio stream 有权
公开(公告)号：US11984127B2
公开(公告)日：2024-05-14
申请号：US17566861
申请日：2021-12-31
申请人： Microsoft Technology Licensing, LLC
发明人： Naoyuki Kanda , Takuya Yoshioka , Zhuo Chen , Jinyu Li , Yashesh Gaur , Zhong Meng , Xiaofei Wang , Xiong Xiao
IPC分类号： G10L15/06 , G10L15/26 , G10L17/04
CPC分类号： G10L17/04 , G10L15/06 , G10L15/26
摘要： The disclosure herein describes using a transcript generation model for generating a transcript from a multi-speaker audio stream. Audio data including overlapping speech of a plurality of speakers is obtained and a set of frame embeddings are generated from audio data frames of the obtained audio data using an audio data encoder. A set of words and channel change (CC) symbols are generated from the set of frame embeddings using a transcript generation model. The CC symbols are included between pairs of adjacent words that are spoken by different people at the same time. The set of words and CC symbols are transformed into a plurality of transcript lines, wherein words of the set of words are sorted into transcript lines based on the CC symbols, and a multi-speaker transcript is generated based on the plurality of transcript lines. The inclusion of CC symbols by the model enables efficient, accurate multi-speaker transcription.

9. 发明授权

US11527238B2 Internal language model for E2E models 有权
公开(公告)号：US11527238B2
公开(公告)日：2022-12-13
申请号：US17154956
申请日：2021-01-21
申请人： Microsoft Technology Licensing, LLC
发明人： Zhong Meng , Sarangarajan Parthasarathy , Xie Sun , Yashesh Gaur , Naoyuki Kanda , Liang Lu , Xie Chen , Rui Zhao , Jinyu Li , Yifan Gong
IPC分类号： G10L15/16 , G06N3/04 , G10L15/01 , G10L15/06 , G10L15/183
摘要： A computer device is provided that includes one or more processors configured to receive an end-to-end (E2E) model that has been trained for automatic speech recognition with training data from a source-domain, and receive an external language model that has been trained with training data from a target-domain. The one or more processors are configured to perform an inference of the probability of an output token sequence given a sequence of input speech features. Performing the inference includes computing an E2E model score, computing an external language model score, and computing an estimated internal language model score for the E2E model. The estimated internal language model score is computed by removing a contribution of an intrinsic acoustic model. The processor is further configured to compute an integrated score based at least on E2E model score, the external language model score, and the estimated internal language model score.

10. 发明授权

US11217265B2 Condition-invariant feature extraction network 有权
公开(公告)号：US11217265B2
公开(公告)日：2022-01-04
申请号：US16434665
申请日：2019-06-07
申请人： Microsoft Technology Licensing, LLC
发明人： Zhong Meng , Yong Zhao , Jinyu Li , Yifan Gong
IPC分类号： G10L25/03 , G10L25/30 , H04R29/00 , G06N20/00 , G06F17/18 , G06N3/08 , H04R5/04
摘要： To generate substantially condition-invariant and speaker-discriminative features, embodiments are associated with a feature extractor capable of extracting features from speech frames based on first parameters, a speaker classifier capable of identifying a speaker based on the features and on second parameters, and a condition classifier capable of identifying a noise condition based on the features and on third parameters. The first parameters of the feature extractor and the second parameters of the speaker classifier are trained to minimize a speaker classification loss, the first parameters of the feature extractor are further trained to maximize a condition classification loss, and the third parameters of the condition classifier are trained to minimize the condition classification loss.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式