利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

53 条记录 (耗时 0.026 秒)

    TECHNIQUES FOR UNIFIED ACOUSTIC ECHO SUPPRESSION USING A RECURRENT NEURAL NETWORK
    21.
    发明公开
    TECHNIQUES FOR UNIFIED ACOUSTIC ECHO SUPPRESSION USING A RECURRENT NEURAL NETWORK 审中-公开

    公开(公告)号：US20230403505A1

    公开(公告)日：2023-12-14

    申请号：US17840188

    申请日：2022-06-14

    申请人： Tencent America LLC

    发明人： Meng Yu ,  Yong Xu ,  Chunlei Zhang ,  Shi-xiong Zhang ,  Dong Yu

    IPC分类号： H04R3/02

    CPC分类号： H04R3/02

    摘要： A method of acoustic echo suppression using a recurrent neural network, performed by at least one processor, is provided. The method includes receiving a microphone signal and a far-end reference signal, estimating an echo suppressed signal and an echo signal based on the microphone signal and the far-end reference signal, estimating enhancement filters for the microphone signal based on the echo suppressed signal and the echo signal, generating an enhanced signal based on the enhancement filters, and adjusting the enhanced signal using automatic gain control (AGC) and outputting the adjusted signal.

    NEURAL-ECHO: AN UNIFIED DEEP NEURAL NETWORK MODEL FOR ACOUSTIC ECHO CANCELLATION AND RESIDUAL ECHO SUPPRESSION
    22.
    发明公开
    NEURAL-ECHO: AN UNIFIED DEEP NEURAL NETWORK MODEL FOR ACOUSTIC ECHO CANCELLATION AND RESIDUAL ECHO SUPPRESSION 审中-公开

    公开(公告)号：US20230395091A1

    公开(公告)日：2023-12-07

    申请号：US18452992

    申请日：2023-08-21

    申请人： TENCENT AMERICA LLC

    发明人： Meng YU ,  Dong Yu

    IPC分类号： G10L21/0224 ,  H04R3/04 ,  G06N3/02

    CPC分类号： G10L21/0224 ,  H04R3/04 ,  G06N3/02 ,  G10L2021/02163

    摘要： A method, computer program, and computer system is provided for an all-deep-learning based AEC system by recurrent neural networks. The model consists of two stages, echo estimation stage and echo suppression stage, respectively. Two different schemes for echo estimation are presented herein: linear echo estimation by multi-tap filtering on far-end reference signal and non-linear echo estimation by single-tap masking on microphone signal. A microphone signal waveform and a far-end reference signal waveform are received. An echo signal waveform is estimated based on the microphone signal waveform and a far-end reference signal waveform. A near-end speech signal waveform is output based on subtracting the estimated echo signal waveform from the microphone signal waveform, and echoes are suppressed within the near-end speech signal waveform.

    Token-wise training for attention based end-to-end speech recognition
    24.
    发明授权
    Token-wise training for attention based end-to-end speech recognition 有权

    公开(公告)号：US11636848B2

    公开(公告)日：2023-04-25

    申请号：US17316856

    申请日：2021-05-11

    申请人： TENCENT AMERICA LLC

    发明人： Peidong Wang ,  Jia Cui ,  Chao Weng ,  Dong Yu

    IPC分类号： G10L15/06 ,  G10L15/22 ,  G10L15/14 ,  G06N20/00 ,  G06N7/01

    摘要： A method of attention-based end-to-end (A-E2E) automatic speech recognition (ASR) training, includes performing cross-entropy training of a model, based on one or more input features of a speech signal, determining a posterior probability vector at a time of a first wrong token among one or more output tokens of the model of which the cross-entropy training is performed, and determining a loss of the first wrong token at the time, based on the determined posterior probability vector. The method further includes determining a total loss of a training set of the model of which the cross-entropy training is performed, based on the determined loss of the first wrong token, and updating the model of which the cross-entropy training is performed, based on the determined total loss of the training set.

    Duration informed attention network for text-to-speech analysis
    26.
    发明授权
    Duration informed attention network for text-to-speech analysis 有权

    公开(公告)号：US11468879B2

    公开(公告)日：2022-10-11

    申请号：US16397349

    申请日：2019-04-29

    申请人： TENCENT AMERICA LLC

    发明人： Chengzhu Yu ,  Heng Lu ,  Dong Yu

    IPC分类号： G10L13/08 ,  G10L13/047 ,  G10L13/00

    摘要： A method and apparatus include receiving a text input that includes a sequence of text components. Respective temporal durations of the text components are determined using a duration model. A first set of spectra is generated based on the sequence of text components. A second set of spectra is generated based on the first set of spectra and the respective temporal durations of the sequence of text components. A spectrogram frame is generated based on the second set of spectra. An audio waveform is generated based on the spectrogram frame. The audio waveform is provided as an output.

    LEARNABLE SPEED CONTROL OF SPEECH SYNTHESIS
    28.
    发明申请
    LEARNABLE SPEED CONTROL OF SPEECH SYNTHESIS 有权

    公开(公告)号：US20220180856A1

    公开(公告)日：2022-06-09

    申请号：US17679790

    申请日：2022-02-24

    申请人： TENCENT AMERICA LLC

    发明人： Chengzhu Yu ,  Dong Yu

    IPC分类号： G10L13/033 ,  G10L25/18 ,  G10L13/06 ,  G10L13/047 ,  G10L25/24

    摘要： A method, computer program, and computer system is provided for synthesizing speech at one or more speeds. A context associated with one or more phonemes corresponding to a speaking voice is encoded, and the one or more phonemes are aligned to one or more target acoustic frames based on the encoded context. One or more mel-spectrogram features are recursively generated from the aligned phonemes and target acoustic frames, and a voice sample corresponding to the speaking voice is synthesized using the generated mel-spectrogram features.

    Singing voice conversion
    30.
    发明授权
    Singing voice conversion 有权

    公开(公告)号：US11183168B2

    公开(公告)日：2021-11-23

    申请号：US16789674

    申请日：2020-02-13

    申请人： TENCENT AMERICA LLC

    发明人： Chengzhu Yu ,  Heng Lu ,  Chao Weng ,  Dong Yu

    IPC分类号： G10L13/027 ,  G10L13/07 ,  G10L13/047 ,  G10L13/00

    摘要： A method, computer program, and computer system is provided for converting a singing first singing voice associated with a first speaker to a second singing voice associated with a second speaker. A context associated with one or more phonemes corresponding to the first singing voice is encoded, and the one or more phonemes are aligned to one or more target acoustic frames based on the encoded context. One or more mel-spectrogram features are recursively generated from the aligned phonemes and target acoustic frames, and a sample corresponding to the first singing voice is converted to a sample corresponding to the second singing voice using the generated mel-spectrogram features.