论文翻译
看过的论文记录
IMU_Pandade
As long as you are here
展开
-
MOCKINGJAY: UNSUPERVISED SPEECH REPRESENTATION LEARNING WITH DEEP BIDIRECTIONAL TRANSFORMER ENCODERS
文章:MOCKINGJAY: UNSUPERVISED SPEECH REPRESENTATION LEARNING WITH DEEP BIDIRECTIONAL TRANSFORMER ENCODERS作者:Andy T. Liu Shu-wen Yang Po-Han Chi Po-chun Hsu Hung-yi LeeNational Taiwan UniversityGitHub:https://github.com/andi611/Self-Supervised-Speech-Pretr原创 2020-09-07 15:51:47 · 996 阅读 · 0 评论 -
(IS 15)Convolutional Neural Networks for Small-footprint Keyword Spotting
会议:INTERSPEECH 2015论文:Convolutional Neural Networks for Small-footprint Keyword Spotting作者:Tara N. Sainath, Carolina ParadaAbstract我们探索使用卷积神经网络(CNN)进行小尺寸关键字发现(KWS)任务。 CNN对于KWS具有吸引力,因为它在参数方面要远远优于DN...原创 2020-05-07 07:54:07 · 715 阅读 · 0 评论 -
(IS 19)On Learning Interpretable CNNs with Parametric Modulated Kernel-based Filters
会议:INTERSPEECH 2019论文:On Learning Interpretable CNNs with Parametric Modulated Kernel-based Filters(基于参数调制的基于核的滤波器学习可解释的CNN)作者:Erfan Loweimi, Peter Bell, Steve RenalsAbstract我们研究了在卷积神经网络(CNN)框架中使用...原创 2020-04-18 11:17:14 · 499 阅读 · 0 评论 -
(IS 19)Feature exploration for almost zero-resource ASR-free keyword spotting using a multilingual b
会议:INTERSPEECH 2019论文:Feature exploration for almost zero-resource ASR-free keyword spotting using a multilingual bottleneck extractor and correspondence autoencoders作者:Raghav Menon, Herman Kamper, ...原创 2020-04-18 11:01:20 · 433 阅读 · 0 评论 -
(IS 19)Automatic Detection of Prosodic Focus in American English
会议:INTERSPEECH 2019论文:Automatic Detection of Prosodic Focus in American English作者:Sunghye Cho, Mark Liberman, Yong-cheol LeeAbstract焦点通常由韵律的突出来调节,突出强调句子中的特定元素以进行强调或对比。尽管它在交流中很重要,但在语音识别领域却很少受到关注。本文...原创 2020-04-16 08:08:36 · 428 阅读 · 0 评论 -
(IS 19)wav2vec: Unsupervised Pre-training for Speech Recognition
会议:INTERSPEECH 2019论文:wav2vec: Unsupervised Pre-training for Speech Recognition作者:Steffen Schneider, Alexei Baevski, Ronan Collobert, Michael AuliAbstract我们通过学习原始音频的表示,探索语音识别的无监督预训练。 在大量未标记的音频数据上对...原创 2020-04-15 18:03:25 · 3480 阅读 · 0 评论 -
(IS 19)Binary Speech Features for Keyword Spotting Tasks Alexandre Riviello, Jean-Pierre David(重点)
会议:INTERSPEECH 2019论文:Binary Speech Features for Keyword Spotting TasksAlexandre Riviello, Jean-Pierre David作者:Alexandre Riviello, Jean-Pierre DavidAbstract关键字发现是一项分类任务,旨在检测一组特定的口语单词。 通常,此类任务在功耗受...原创 2020-04-15 08:10:23 · 453 阅读 · 0 评论 -
(IS 19)Low-Dimensional Bottleneck Features for On-Device Continuous Speech Recognition
会议:INTERSPEECH 2019论文:Low-Dimensional Bottleneck Features for On-Device Continuous Speech Recognition作者:David B. Ramsay, Kevin Kilgour, Dominik Roblek, Matthew SharifiAbstract低功耗数字信号处理器(DSP)通常具有非常...原创 2020-04-13 09:55:51 · 325 阅读 · 0 评论 -
(IS 19)Unsupervised Raw Waveform Representation Learning for ASR
会议:INTERSPEECH 2019论文:Unsupervised Raw Waveform Representation Learning for ASR作者:Purvi Agrawal, Sriram GanapathyAbstract在本文中,我们提出了一种在无监督学习范例中使用原始语音波形的深度表示学习方法。提出的深度模型的第一层执行声学滤波,而随后的一层执行调制滤波。使用学习其...原创 2020-04-08 16:50:50 · 595 阅读 · 0 评论 -
(IS 19)Prosody Usage Optimization for Children Speech Recognition with Zero Resource Children Speech
会议:INTERSPEECH 2019论文:Prosody Usage Optimization for Children Speech Recognition with Zero Resource Children Speech作者:Chenda Li, Yanmin QianAbstract儿童语音识别仍然是自动语音识别的一大挑战。由于处理过程更加困难且数据收集成本较高,因此大多数当前...原创 2020-04-03 08:20:52 · 399 阅读 · 0 评论 -
(IS 19)Modulation Vectors as Robust Feature Representation for ASR in Domain Mismatched Conditions
会议:INTERSPEECH 2019论文:Modulation Vectors as Robust Feature Representation for ASR in Domain Mismatched Conditions作者:Samik Sadhu, Hynek HermanskyAbstract在这项工作中,我们在自动语音识别(ASR)系统中的训练和测试条件之间的域不匹配中,证明了...原创 2020-04-03 07:48:33 · 415 阅读 · 0 评论 -
(2015)Deep Residual Learning for Image Recognition
会议:CVPR, 2016, pp. 770–778.论文:Deep Residual Learning for Image Recognition作者:Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun原创 2020-03-24 16:03:59 · 549 阅读 · 0 评论 -
(2017)Honk: A PyTorch Reimplementation of Convolutional Neural Networks for Keyword Spotting
论文:Honk: A PyTorch Reimplementation of Convolutional Neural Networks for Keyword Spotting作者:Raphael Tang, Jimmy LinABSTRACT我们描述了Honk,这是TensorFlow示例中包含的用于关键字识别的卷积神经网络的开源PyTorch重新实现。 这些模型对于识别基于语音的界面(...原创 2020-03-24 16:04:52 · 691 阅读 · 0 评论 -
(ICASSP 18)DEEP RESIDUAL LEARNING FOR SMALL-FOOTPRINT KEYWORD SPOTTING(重点)
会议:ICASSP 2018论文:DEEP RESIDUAL LEARNING FOR SMALL-FOOTPRINT KEYWORD SPOTTING、链接2、GitHub作者:Raphael Tang ; Jimmy Lin原创 2020-03-24 16:05:08 · 1357 阅读 · 0 评论 -
(ICASSP 19)Streaming End-to-end Speech Recognition for Mobile Devices
会议:ICASSP 2019论文:Streaming End-to-end Speech Recognition for Mobile Devices作者:Yanzhang He, Tara N. Sainath, Rohit Prabhavalkar, Ian McGraw, Raziel Alvarez, Ding Zhao, David Rybach, Anjuli Kannan, Yo...原创 2020-03-24 16:05:50 · 1834 阅读 · 0 评论 -
(ICASSP 19)AUTOMATIC GRAMMAR AUGMENTATION FOR ROBUST VOICE COMMAND RECOGNITION
会议:ICASSP 2019论文:作者:原创 2020-03-24 16:06:07 · 588 阅读 · 0 评论 -
(ICASSP 19)SEMI-SUPERVISED AND POPULATION BASED TRAINING FOR VOICE COMMANDS(Speech Commands Dataset)
会议:ICASSP 2019论文:SEMI-SUPERVISED AND POPULATION BASED TRAINING FOR VOICE COMMANDS RECOGNITION作者:Oguz H. Elibol ; Gokce Keskin ; Anil ThomasAbstract提出了一种将超参数自动调整与半监督训练相结合的快速设计方法,建立了高精度、鲁棒的语音命令分类模型。...原创 2020-03-24 16:06:34 · 765 阅读 · 0 评论 -
(ICASSP 19)FOCAL LOSS AND DOUBLE-EDGE-TRIGGERED DETECTOR FOR ROBUST SMALL-FOOTPRINT KEYWORD SPOTTING
会议:ICASSP 2019论文:FOCAL LOSS AND DOUBLE-EDGE-TRIGGERED DETECTOR FOR ROBUST SMALL-FOOTPRINT KEYWORD SPOTTING作者:Bin Liu ; Shuai Nie ; Yaping Zhang ; Shan Liang ; Zhanlei Yang ; Wenju LiuABSTRACT关键词识别...原创 2020-03-24 16:06:53 · 794 阅读 · 0 评论 -
(ICASSP 19)ADVERSARIAL EXAMPLES FOR IMPROVING END-TO-END ATTENTION-BASED SMALL-FOOTPRINT KEYWORD SPO
会议:ICASSP 2019论文:ADVERSARIAL EXAMPLES FOR IMPROVING END-TO-END ATTENTION-BASED SMALL-FOOTPRINT KEYWORD SPOTTING作者:Xiong Wang ; Sining Sun ; Changhao Shan ; Jingyong Hou ; Lei Xie ; Shen Li ; Xin Lei...原创 2020-03-24 16:07:31 · 502 阅读 · 0 评论 -
(ICASSP 19)VOICE TRIGGER DETECTION FROM LVCSR HYPOTHESIS LATTICES USING BIDIRECTIONAL LATTICE RECURR
会议:ICASSP 2019论文:VOICE TRIGGER DETECTION FROM LVCSR HYPOTHESIS LATTICES USINGBIDIRECTIONAL LATTICE RECURRENT NEURAL NETWORKS作者:Woojay Jeon ; Leo Liu ; Henry MasonABSTRACT我们提出了一种通过神经网络对服务器端大型词汇连续语...原创 2020-03-24 16:07:49 · 370 阅读 · 0 评论 -
(ICASSP 19)Hotword Cleaner: Dual-microphone Adaptive Noise Cancellation with Deferred Filter Coeffic
会议:ICASSP 2019论文:Hotword Cleaner: Dual-microphone Adaptive Noise Cancellation with Deferred Filter Coefficients for Robust Keyword Spotting作者:Yiteng Arden Huang ; Turaj Z. Shabestary ; Alexander Gru...原创 2020-03-24 16:08:42 · 784 阅读 · 0 评论 -
(ICASSP 18)Temporal Modeling Using Dilated Convolution and Gating for Voice-Activity-Detection
会议:ICASSP 2018论文:Temporal Modeling Using Dilated Convolution and Gating for Voice-Activity-Detection作者:Shuo-Yiin Chang, Bo Li, Gabor Simko, Tara N Sainath, Anshuman Tripathi, Aäron van den Oord, Ori...原创 2020-03-24 16:09:02 · 746 阅读 · 0 评论 -
(KWS-LSTM)Max-pooling loss training of long short-term memory networks for small-footprint keyword s
会议:2016 IEEE口语技术研讨会(SLT)论文:Max-pooling loss training of long short-term memory networks for small-footprint keyword spotting作者: Ming Ming,Anirudh Raju,George Tucker,Sankaran Panchapagesan,Gengshen F...原创 2020-03-24 16:09:19 · 1474 阅读 · 0 评论 -
(Interspeech 15)Convolutional neural networks for small-footprint keyword spotting
会议:Sixteenth Annual Conference of the International Speech Communication Association, 2015.论文:“Convolutional neural networks for small-footprint keyword spotting”,作者:Tara N Sainath, Carolina Parada...原创 2020-03-24 16:03:38 · 599 阅读 · 0 评论 -
(ICASSP 2014)Small-footprint keyword spotting using deep neural networks
会议:ICASSP 2014论文:Small-footprint keyword spotting using deep neural networks作者:Guoguo Chen ; Carolina Parada ; Georg HeigoldAbstract我们的应用程序需要具有内存占用量小,计算成本低和精度高的关键字查找系统。为了满足这些要求,我们提出了一种基于深度神经网络的简单方...原创 2020-03-24 16:10:43 · 1579 阅读 · 0 评论 -
(KWS-HMM)
会议:ICASSP-90论文:A HIDDEN MARKOV MODEL BASED KEYWORD RECOGNITION SYSTEM作者:Richard C Rose,Douglas B Paul原创 2020-03-24 16:11:21 · 803 阅读 · 0 评论 -
(ICASSP 19)EFFICIENT KEYWORD SPOTTING USING DILATED CONVOLUTIONS AND GATING
会议:ICASSP 2019论文:EFFICIENT KEYWORD SPOTTING USING DILATED CONVOLUTIONS AND GATING作者:Alice Coucke, Mohammed Chlieh, Thibault Gisselbrecht, David Leroy,Mathieu Poumeyrol, Thibaut LavrilABSTRACT我们探索...原创 2020-03-24 16:11:37 · 1028 阅读 · 1 评论 -
(ISCSLP 16)End-to-end keywords spotting based on connectionist temporal classification for Mandarin
会议:ISCSLP 2016论文:作者:原创 2020-03-24 16:14:40 · 970 阅读 · 0 评论 -
(ICACSIS 17)Contextual keyword spotting in lecture video with deep convolutional neural network
论文:Contextual keyword spotting in lecture video with deep convolutional neural network发表于: 2017年高级计算机科学与信息系统国际会议(ICACSIS)加入IEEE Xplore的日期: 2018年5月7日Abstract 介绍了使用深度卷积神经网络(CNN)架构的演讲视频关键字发现(KWS)系统。...原创 2020-03-24 16:14:23 · 576 阅读 · 0 评论 -
(ICASSP 19)Federated Learning for Keyword Spotting
会议: ICASSP 2019论文:Federated Learning for Keyword Spotting作者:David Leroy、Alice Coucke、Thibaut Lavril、Thibault Gisselbrecht、Joseph DureauABSTRACT提出了一种基于联合学习的实用方法,以通过连续运行基于嵌入式语音的模型(例如唤醒词检测器)来解决域外问题。我...原创 2020-03-24 16:14:56 · 1038 阅读 · 0 评论 -
(IEEE Access7)Effective Combination of DenseNet and BiLSTM for Keyword Spotting
论文地址:Effective Combination of DenseNet and BiLSTM for Keyword Spotting发表于: IEEE Access ( 第7卷)发布日期: 2019年1月10日Abstract 在本文中,基于DenseNet提取本地特征图的强大功能,我们为KWS提出了一种新的网络体系结构(DenseNet-BiLSTM)。在我们的DenseNet...原创 2020-03-24 16:15:57 · 1230 阅读 · 0 评论 -
(INTERSPEECH 19)Full-Sentence Correlation: a Method to Handle Unpredictable Noise for Robust Speech
会议:INTERSPEECH 2019论文:Full-Sentence Correlation: a Method to Handle Unpredictable Noise for Robust Speech Recognition作者:Ji Ming, Danny CrookesAbstract 描述了用于语音识别的全句语音相关的理论和实现,并证明了它对未经训练/未经训练的噪声具有优...原创 2020-03-24 16:16:30 · 453 阅读 · 0 评论 -
(ICASSP 19)END-TO-END STREAMING KEYWORD SPOTTING
会议:ICASSP 2019论文:END-TO-END STREAMING KEYWORD SPOTTING作者:Raziel Alvarez, Hyun Jin Park, Google, Inc., United StatesABSTRACT 提出了一个关键词识别系统,除了用于特征生成的前端组件外,它完全包含在经过“端到端”训练的深度神经网络(DNN)模型中,用于预测音频流中关键词的...原创 2020-03-24 16:17:33 · 1436 阅读 · 1 评论 -
论文笔记 ---语音关键词检测方法综述
概述相比于语音识别、语音合成、语音增强,说话人识别等常见语音领域,关键词检测相对来说比较小众,但随着智能助理、智能音箱等的兴起,关键词检测越来越受到产业界的 重视。语音关键词检测关注如何从连续语音流中检测出用户感兴趣的关键词。典型场景分为两类:1、语音设备控制: 根据用户指令来唤醒或者控制智能设备;2、语音检索: 从大段语音文档中定位到关键词所在位置。Keyword Spotting 指...原创 2020-03-24 16:17:45 · 3279 阅读 · 0 评论 -
论文笔记 - 《A Fast Learning Algorithm for Deep Belief Net》---深度学习前夕
Hinton, Geoffrey E., Simon Osindero, and Yee-Whye Teh. “A fast learning algorithm for deep belief nets.” Neural computation 18.7 (2006): 1527-1554. [pdf](Deep Learning Eve)作者: G.E.Hinton et. al.日期: ...原创 2020-03-24 16:17:58 · 1140 阅读 · 0 评论 -
论文笔记 - 《Deep Learning》(Yann LeCun Yoshua Bengio & Geoffrey Hinton)经典
论文: LeCun, Yann, Yoshua Bengio, and Geoffrey Hinton. “Deep learning.” Nature 521.7553 (2015): 436-444. [pdf] (Three Giants’ Survey)监督学习机器学习最常见的形式,不管是否深入都是监督学习。我们计算一个目标函数,它度量输出分数与期望的分数模式之间的误差(距离)。然...原创 2020-03-24 16:18:07 · 2379 阅读 · 0 评论