公开(公告)号：US20240362453A1

公开(公告)日：2024-10-31

申请号：US18766038

申请日：2024-07-08

申请人： Google LLC

发明人： Anmol Gulati ,  Weikeng Qin ,  Zhengdong Zhang ,  Ruoming Pang ,  Niki Parmar ,  Jiahui Yu ,  Wei Han ,  Chung-Cheng Chiu ,  Yu Zhang ,  Yonghui Wu ,  Shibo Wang

IPC分类号： G06N3/04 ,  G06N20/00 ,  G10L15/16

CPC分类号： G06N3/04 ,  G06N20/00 ,  G10L15/16

摘要： Systems and methods can utilize a conformer model to process a data set for various data processing tasks, including, but not limited to, speech recognition, sound separation, protein synthesis determination, video or other image set analysis, and natural language processing. The conformer model can use feed-forward blocks, a self-attention block, and a convolution block to process data to learn global interactions and relative-offset-based local correlations of the input data.

公开(公告)号：US12079703B2

公开(公告)日：2024-09-03

申请号：US17139525

申请日：2020-12-31

申请人： Google LLC

发明人： Anmol Gulati ,  Ruoming Pang ,  Niki Parmar ,  Jiahui Yu ,  Wei Han ,  Chung-Cheng Chiu ,  Yu Zhang ,  Yonghui Wu ,  Shibo Wang ,  Weikeng Qin ,  Zhengdong Zhang

IPC分类号： G06N3/04 ,  G06N20/00 ,  G10L15/16

CPC分类号： G06N3/04 ,  G06N20/00 ,  G10L15/16

摘要： Systems and methods can utilize a conformer model to process a data set for various data processing tasks, including, but not limited to, speech recognition, sound separation, protein synthesis determination, video or other image set analysis, and natural language processing. The conformer model can use feed-forward blocks, a self-attention block, and a convolution block to process data to learn global interactions and relative-offset-based local correlations of the input data.

公开(公告)号：US12032920B2

公开(公告)日：2024-07-09

申请号：US17056554

申请日：2020-03-07

申请人： Google LLC

发明人： Ye Jia ,  Zhifeng Chen ,  Yonghui Wu ,  Melvin Johnson ,  Fadi Biadsy ,  Ron Weiss ,  Wolfgang Macherey

IPC分类号： G06F40/47 ,  G06F40/58

CPC分类号： G06F40/47 ,  G06F40/58

摘要： The present disclosure provides systems and methods that train and use machine-learned models such as, for example, sequence-to-sequence models, to perform direct and text-free speech-to-speech translation. In particular, aspects of the present disclosure provide an attention-based sequence-to-sequence neural network which can directly translate speech from one language into speech in another language, without relying on an intermediate text representation.

公开(公告)号：US20240161732A1

公开(公告)日：2024-05-16

申请号：US18418246

申请日：2024-01-20

申请人： Google LLC

发明人： Zhifeng Chen ,  Bo Li ,  Eugene Weinstein ,  Yonghui Wu ,  Pedro J. Moreno Mengibar ,  Ron J. Weiss ,  Khe Chai Sim ,  Tara N. Sainath ,  Patrick An Phu Nguyen

IPC分类号： G10L15/00 ,  G10L15/07 ,  G10L15/16

CPC分类号： G10L15/005 ,  G10L15/07 ,  G10L15/16 ,  G10L2015/0631

摘要： Methods, systems, and apparatus, including computer programs encoded on a computer-readable media, for speech recognition using multi-dialect and multilingual models. In some implementations, audio data indicating audio characteristics of an utterance is received. Input features determined based on the audio data are provided to a speech recognition model that has been trained to output score indicating the likelihood of linguistic units for each of multiple different language or dialects. The speech recognition model can be one that has been trained using cluster adaptive training. Output that the speech recognition model generated in response to receiving the input features determined based on the audio data is received. A transcription of the utterance generated based on the output of the speech recognition model is provided.

公开(公告)号：US20240112667A1

公开(公告)日：2024-04-04

申请号：US18525475

申请日：2023-11-30

申请人： Google LLC

发明人： Ye Jia ,  Zhifeng Chen ,  Yonghui Wu ,  Jonathan Shen ,  Ruoming Pang ,  Ron J. Weiss ,  Ignacio Lopez Moreno ,  Fei Ren ,  Yu Zhang ,  Quan Wang ,  Patrick An Phu Nguyen

IPC分类号： G10L13/04 ,  G10L17/04 ,  G10L19/00

CPC分类号： G10L13/04 ,  G10L17/04 ,  G10L19/00 ,  G06N3/08

摘要： Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for speech synthesis. The methods, systems, and apparatus include actions of obtaining an audio representation of speech of a target speaker, obtaining input text for which speech is to be synthesized in a voice of the target speaker, generating a speaker vector by providing the audio representation to a speaker encoder engine that is trained to distinguish speakers from one another, generating an audio representation of the input text spoken in the voice of the target speaker by providing the input text and the speaker vector to a spectrogram generation engine that is trained using voices of reference speakers to generate audio representations, and providing the audio representation of the input text spoken in the voice of the target speaker for output.

公开(公告)号：US11922932B2

公开(公告)日：2024-03-05

申请号：US18194586

申请日：2023-03-31

申请人： Google LLC

发明人： Rohit Prakash Prabhavalkar ,  Tara N. Sainath ,  Yonghui Wu ,  Patrick An Phu Nguyen ,  Zhifeng Chen ,  Chung-Cheng Chiu ,  Anjuli Patricia Kannan

IPC分类号： G10L15/197 ,  G10L15/02 ,  G10L15/06 ,  G10L15/16 ,  G10L15/22

CPC分类号： G10L15/197 ,  G10L15/02 ,  G10L15/063 ,  G10L15/16 ,  G10L15/22 ,  G10L2015/025

摘要： Methods, systems, and apparatus, including computer programs encoded on computer-readable storage media, for speech recognition using attention-based sequence-to-sequence models. In some implementations, audio data indicating acoustic characteristics of an utterance is received. A sequence of feature vectors indicative of the acoustic characteristics of the utterance is generated. The sequence of feature vectors is processed using a speech recognition model that has been trained using a loss function that uses a set of speech recognition hypothesis samples, the speech recognition model including an encoder, an attention module, and a decoder. The encoder and decoder each include one or more recurrent neural network layers. A sequence of output vectors representing distributions over a predetermined set of linguistic units is obtained. A transcription for the utterance is obtained based on the sequence of output vectors. Data indicating the transcription of the utterance is provided.

公开(公告)号：US11908448B2

公开(公告)日：2024-02-20

申请号：US17327076

申请日：2021-05-21

申请人： Google LLC

发明人： Isaac Elias ,  Jonathan Shen ,  Yu Zhang ,  Ye Jia ,  Ron J. Weiss ,  Yonghui Wu ,  Byungha Chun

IPC分类号： G10L13/08 ,  G10L13/047 ,  G06F40/126 ,  G10L21/10 ,  G06N3/08 ,  G06N3/088 ,  G06N3/044 ,  G06N3/045 ,  G06N3/048

CPC分类号： G10L13/08 ,  G06F40/126 ,  G06N3/044 ,  G06N3/045 ,  G06N3/08 ,  G06N3/088 ,  G10L13/047 ,  G10L21/10 ,  G06N3/048

摘要： A method for training a non-autoregressive TTS model includes receiving training data that includes a reference audio signal and a corresponding input text sequence. The method also includes encoding the reference audio signal into a variational embedding that disentangles the style/prosody information from the reference audio signal and encoding the input text sequence into an encoded text sequence. The method also includes predicting a phoneme duration for each phoneme in the input text sequence and determining a phoneme duration loss based on the predicted phoneme durations and a reference phoneme duration. The method also includes generating one or more predicted mel-frequency spectrogram sequences for the input text sequence and determining a final spectrogram loss based on the predicted mel-frequency spectrogram sequences and a reference mel-frequency spectrogram sequence. The method also includes training the TTS model based on the final spectrogram loss and the corresponding phoneme duration loss.

公开(公告)号：US11556381B2

公开(公告)日：2023-01-17

申请号：US17738909

申请日：2022-05-06

申请人： Google LLC

发明人： Jeffrey Adgate Dean ,  Sudip Roy ,  Michael Acheson Isard ,  Aakanksha Chowdhery ,  Brennan Saeta ,  Chandramohan Amyangot Thekkath ,  Daniel William Hurt ,  Hyeontaek Lim ,  Laurent El Shafey ,  Parker Edward Schuh ,  Paul Ronald Barham ,  Ruoming Pang ,  Ryan Sepassi ,  Sanjay Ghemawat ,  Yonghui Wu

IPC分类号： G06F15/82 ,  G06F9/48 ,  G06N3/08 ,  G06N3/063

摘要： Methods, systems, and apparatus, including computer programs encoded on computer storage media, for distributing machine learning workloads, e.g., computations for training a neural network or computing an inference using a neural network, across multiple hardware accelerators. One of the systems comprises a plurality of accelerator islands, each hardware accelerator island comprising a respective plurality of hardware devices that include a plurality of hardware accelerators and a corresponding host for each of the plurality of hardware accelerators; and a respective scheduler for each of the accelerator islands that is configured to schedule workloads across the plurality of accelerators and corresponding hosts in the accelerator island, wherein the system is configured to: receive data representing a machine learning workload; and assign a respective portion of the machine learning workload to each of the plurality of accelerator islands for scheduling by the respective scheduler for the accelerator island.

公开(公告)号：US11475874B2

公开(公告)日：2022-10-18

申请号：US17163007

申请日：2021-01-29

申请人： Google LLC

发明人： Yu Zhang ,  Bhuvana Ramabhadran ,  Andrew Rosenberg ,  Yonghui Wu ,  Byungha Chun ,  Ron Weiss ,  Yuan Cao

IPC分类号： G10L25/30 ,  G10L25/00 ,  G10L17/00 ,  G10L13/047 ,  G10L25/18 ,  G06N3/08 ,  G10L15/06 ,  G10L13/10

摘要： A method of generating diverse and natural text-to-speech (TTS) samples includes receiving a text and generating a speech sample based on the text using a TTS model. A training process trains the TTS model to generate the speech sample by receiving training samples. Each training sample includes a spectrogram and a training text corresponding to the spectrogram. For each training sample, the training process identifies speech units associated with the training text. For each speech unit, the training process generates a speech embedding, aligns the speech embedding with a portion of the spectrogram, extracts a latent feature from the aligned portion of the spectrogram, and assigns a quantized embedding to the latent feature. The training process generates the speech sample by decoding a concatenation of the speech embeddings and a quantized embeddings for the speech units associated with the training text corresponding to the spectrogram.

公开(公告)号：US11468244B2

公开(公告)日：2022-10-11

申请号：US16834342

申请日：2020-03-30

申请人： Google LLC

发明人： Anjuli Patricia Kannan ,  Tara N. Sainath ,  Yonghui Wu ,  Ankur Bapna ,  Arindrima Datta

IPC分类号： G10L15/00 ,  G06F40/40

摘要： A method of transcribing speech using a multilingual end-to-end (E2E) speech recognition model includes receiving audio data for an utterance spoken in a particular native language, obtaining a language vector identifying the particular language, and processing, using the multilingual E2E speech recognition model, the language vector and acoustic features derived from the audio data to generate a transcription for the utterance. The multilingual E2E speech recognition model includes a plurality of language-specific adaptor modules that include one or more adaptor modules specific to the particular native language and one or more other adaptor modules specific to at least one other native language different than the particular native language. The method also includes providing the transcription for output.