WO2020255242A1 - 復元装置、復元方法、およびプログラム - Google Patents

復元装置、復元方法、およびプログラム Download PDF

Info

Publication number
WO2020255242A1
WO2020255242A1 PCT/JP2019/024058 JP2019024058W WO2020255242A1 WO 2020255242 A1 WO2020255242 A1 WO 2020255242A1 JP 2019024058 W JP2019024058 W JP 2019024058W WO 2020255242 A1 WO2020255242 A1 WO 2020255242A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
clip
restoration
neural network
post
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2019/024058
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
江村　暁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NTT Inc
Original Assignee
Nippon Telegraph and Telephone Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nippon Telegraph and Telephone Corp filed Critical Nippon Telegraph and Telephone Corp
Priority to JP2021528089A priority Critical patent/JP7188589B2/ja
Priority to US17/619,618 priority patent/US20220375489A1/en
Priority to PCT/JP2019/024058 priority patent/WO2020255242A1/ja
Publication of WO2020255242A1 publication Critical patent/WO2020255242A1/ja
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
    • G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • G10L21/0208—Noise filtering
    • G10L21/0216—Noise filtering characterised by the method used for estimating noise
    • G10L21/0232—Processing in the frequency domain
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/09—Supervised learning

Definitions

  • FIG. 1 is a diagram illustrating a functional configuration of a waveform restoration device.
  • FIG. 2 is a diagram illustrating the configuration of the waveform restoration unit.
  • FIG. 3 is a diagram illustrating a processing procedure of the waveform restoration method.
  • FIG. 4 is a diagram illustrating the functional configuration of the waveform restoration unit of the second embodiment.
  • FIG. 5 is a diagram illustrating a functional configuration of a computer.
  • the signal restoration device of the first embodiment (hereinafter, also referred to as “restoration device”) is a signal restoration neural network composed of a gated convolutional neural network (see, for example, References 1 and 2). , A signal processing device that restores the signal before clipping from the signal after clipping. Since the operation of the neural network is fixed, the total amount of operation of the signal restoration process by the signal restoration neural network is constant. Further, by sufficiently learning the signal restoration neural network in advance using sufficient training data, it can be expected that the characteristics of the signal before clipping are better reflected in the signal after restoration.
  • step S13 the frame synthesizing unit 13 applies the frame synthesizing process to the vector of the estimated pre-clip signal to restore the pre-clip signal.
  • the configuration of the second embodiment can be applied even when the missing signal is restored.
  • the input data is an L ⁇ 2 matrix composed of a missing signal vector and a missing information vector.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Signal Processing (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Quality & Reliability (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)
  • Image Processing (AREA)
PCT/JP2019/024058 2019-06-18 2019-06-18 復元装置、復元方法、およびプログラム Ceased WO2020255242A1 (ja)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2021528089A JP7188589B2 (ja) 2019-06-18 2019-06-18 復元装置、復元方法、およびプログラム
US17/619,618 US20220375489A1 (en) 2019-06-18 2019-06-18 Restoring apparatus, restoring method, and program
PCT/JP2019/024058 WO2020255242A1 (ja) 2019-06-18 2019-06-18 復元装置、復元方法、およびプログラム

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2019/024058 WO2020255242A1 (ja) 2019-06-18 2019-06-18 復元装置、復元方法、およびプログラム

Publications (1)

Publication Number Publication Date
WO2020255242A1 true WO2020255242A1 (ja) 2020-12-24

Family

ID=74037011

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2019/024058 Ceased WO2020255242A1 (ja) 2019-06-18 2019-06-18 復元装置、復元方法、およびプログラム

Country Status (3)

Country Link
US (1) US20220375489A1 (https=)
JP (1) JP7188589B2 (https=)
WO (1) WO2020255242A1 (https=)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023279722A1 (en) 2021-07-06 2023-01-12 Huawei Technologies Co.,Ltd. Method and device for reducing peak-to-average power ration for single carrier signals

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005275410A (ja) * 2004-03-23 2005-10-06 Herman Becker Automotive Systems-Wavemakers Inc ニューラルネットワークを利用してスピーチ信号を分離する。
JP2013162347A (ja) * 2012-02-06 2013-08-19 Sony Corp 画像処理装置、画像処理方法、プログラム、および装置

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150032449A1 (en) * 2013-07-26 2015-01-29 Nuance Communications, Inc. Method and Apparatus for Using Convolutional Neural Networks in Speech Recognition
KR102565447B1 (ko) * 2017-07-26 2023-08-08 삼성전자주식회사 청각 인지 속성에 기반하여 디지털 오디오 신호의 이득을 조정하는 전자 장치 및 방법
US10699700B2 (en) * 2018-07-31 2020-06-30 Tencent Technology (Shenzhen) Company Limited Monaural multi-talker speech recognition with attention mechanism and gated convolutional networks
US20190149134A1 (en) * 2019-01-14 2019-05-16 Intel Corporation Filter optimization to improve computational efficiency of convolution operations

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005275410A (ja) * 2004-03-23 2005-10-06 Herman Becker Automotive Systems-Wavemakers Inc ニューラルネットワークを利用してスピーチ信号を分離する。
JP2013162347A (ja) * 2012-02-06 2013-08-19 Sony Corp 画像処理装置、画像処理方法、プログラム、および装置

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
JIAHUI YU ET AL.: "Free-Form Image Inpainting with Gated Convolution", ARXIV, 10 June 2018 (2018-06-10), pages 1 - 12, XP033723862, Retrieved from the Internet <URL:https://arxiv.org/pdf/1806.03589.pdf> [retrieved on 20190911] *
SATOSHI IIZUKA ET AL., GLOBALLY AND LOCALLY CONSISTENT IMAGE COMPLETION, July 2017 (2017-07-01), pages 1 - 14, XP058372881, Retrieved from the Internet <URL:http://iizuka.cs.tsukuba.ac.jp/projects/completion/data/completion_sig2017.pdf> [retrieved on 20190911] *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023279722A1 (en) 2021-07-06 2023-01-12 Huawei Technologies Co.,Ltd. Method and device for reducing peak-to-average power ration for single carrier signals
EP4352931A4 (en) * 2021-07-06 2024-10-16 Huawei Technologies Co., Ltd. METHOD AND APPARATUS FOR REDUCING THE PEAK-TO-AVERAGE POWER RATIO FOR SINGLE-CARRIER SIGNALS

Also Published As

Publication number Publication date
JP7188589B2 (ja) 2022-12-13
JPWO2020255242A1 (https=) 2020-12-24
US20220375489A1 (en) 2022-11-24

Similar Documents

Publication Publication Date Title
US12530876B2 (en) Systems and methods for progressive learning for machine-learned models to optimize training speed
Afonso et al. An augmented Lagrangian approach to the constrained optimization formulation of imaging inverse problems
Wei et al. Deep unfolding with normalizing flow priors for inverse problems
Adam et al. Image denoising using combined higher order non-convex total variation with overlapping group sparsity
US20200342046A1 (en) Reduced dot product computation circuit
KR20180072562A (ko) 인공 뉴럴 네트워크 클래스-기반 프루닝
JP2021524973A (ja) パタン認識装置、パタン認識方法、及びプログラム
Jiang et al. A data-driven high-resolution time-frequency distribution
WO2020022498A1 (ja) クラスタリング装置、方法、及びプログラム
Li et al. A novel weighted anisotropic total variational model for image applications
Zhang et al. High‐Order Total Bounded Variation Model and Its Fast Algorithm for Poissonian Image Restoration
CN113496228A (zh) 一种基于Res2Net、TransUNet和协同注意力的人体语义分割方法
US20150046377A1 (en) Joint Sound Model Generation Techniques
WO2020255242A1 (ja) 復元装置、復元方法、およびプログラム
WO2020040007A1 (ja) 学習装置、学習方法及び学習プログラム
Zhou et al. Efficient cascaded multiscale adaptive network for image restoration
CN120103126A (zh) 高压真空断路器的故障诊断方法、装置、设备、存储介质和程序产品
CN112749845A (zh) 模型训练方法、资源数据预测方法、装置和计算设备
Wei et al. Identification and reconstruction of chaotic systems using multiresolution wavelet decompositions
Taleb et al. Multiresolution analysis of point processes and statistical thresholding for Haar wavelet-based intensity estimation
De Luca et al. A gpu-based algorithm for environmental data filtering
Kumar et al. VLSI implementation for noise suppression using parallel median filtering technique
Amat et al. On a nonlinear 4-point quaternary approximating subdivision scheme eliminating the Gibbs phenomenon
CN118038355A (zh) 基于多尺度的人群计数模型训练方法、装置及存储介质
Perschewski et al. Pursuing the perfect projection: a projection pursuit framework for deep learning

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19933305

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021528089

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19933305

Country of ref document: EP

Kind code of ref document: A1