公开(公告)号：US20190318228A1

公开(公告)日：2019-10-17

申请号：US16260637

申请日：2019-01-29

Applicant: Electronics and Telecommunications Research Institute

Inventor： Hyun Woo KIM ,  Ho Young JUNG ,  Jeon Gue PARK ,  Yun Keun LEE

IPC: G06N3/04 ,  G06N3/08 ,  G06F3/06 ,  H03H17/02

Abstract: Provided are an apparatus and method for a statistical memory network. The apparatus includes a stochastic memory, an uncertainty estimator configured to estimate uncertainty information of external input signals from the input signals and provide the uncertainty information of the input signals, a writing controller configured to generate parameters for writing in the stochastic memory using the external input signals and the uncertainty information and generate additional statistics by converting statistics of the external input signals, a writing probability calculator configured to calculate a probability of a writing position of the stochastic memory using the parameters for writing, and a statistic updater configured to update stochastic values composed of an average and a variance of signals in the stochastic memory using the probability of a writing position, the parameters for writing, and the additional statistics.

公开(公告)号：US20180165578A1

公开(公告)日：2018-06-14

申请号：US15478342

申请日：2017-04-04

Applicant: Electronics and Telecommunications Research Institute

Inventor： Hoon CHUNG ,  Jeon Gue PARK ,  Sung Joo LEE ,  Yun Keun LEE

IPC: G06N3/08 ,  G06F17/16 ,  G06N3/04

CPC classification number: G06N3/04 ,  G06N3/0481 ,  G06N3/063

Abstract: Provided are an apparatus and method for compressing a deep neural network (DNN). The DNN compression method includes receiving a matrix of a hidden layer or an output layer of a DNN, calculating a matrix representing a nonlinear structure of the hidden layer or the output layer, and decomposing the matrix of the hidden layer or the output layer using a constraint imposed by the matrix representing the nonlinear structure.

公开(公告)号：US20170206894A1

公开(公告)日：2017-07-20

申请号：US15187581

申请日：2016-06-20

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Byung Ok KANG ,  Jeon Gue PARK ,  Hwa Jeon SONG ,  Yun Keun LEE ,  Eui Sok CHUNG

IPC: G10L15/16 ,  G10L15/07

CPC classification number: G10L15/16 ,  G10L15/063 ,  G10L15/07 ,  G10L2015/022 ,  G10L2015/0636

Abstract: A speech recognition apparatus based on a deep-neural-network (DNN) sound model includes a memory and a processor. As the processor executes a program stored in the memory, the processor generates sound-model state sets corresponding to a plurality of pieces of set training speech data included in multi-set training speech data, generates a multi-set state cluster from the sound-model state sets, and sets the multi-set training speech data as an input node and the multi-set state cluster as output nodes so as to learn a DNN structured parameter.