End- To- End Speech Recogni0on with Recurrent Neural Networks
|
|
|
- Anna Garrison
- 5 years ago
- Views:
Transcription
1 RTTH Summer School on Speech Technology: A Deep Learning Perspec0ve End- To- End Speech Recogni0on with Recurrent Neural Networks José A. R. Fonollosa Universitat Politècnica de Catalunya. Barcelona Barcelona, July 9, 2015
2 From speech processing to machine learning 2
3 Towards end- to- end RNN Speech Recogni0on? Architectures GMM- HMM: 30 years of feature engineering DNN- GMM- HMM: Trained features DNN- HMM: TDNN, LSTM, RNN, MS DNN for language modeling (RNN) End- to- end DNN? Examples Alex Graves (Google) Deep Speech (Baidu) Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
4 GMM- HMM Perceptual Feature Extrac0on (MFCC, PLP, FF, VTLN,GammaTone,..) Feature Transforma0on (Deriva0ve, LDA, MLLT, fmllr,..) GMM (Training: ML, MMI, MPE, MWE, SAT,..) HMM N- GRAM Acous&c Model Phone0c inventory Pronuncia0on Lexicon Language Model Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
5 DNN- GMM- HMM Feature Extrac0on (MFCC, PLP, FF) DNN (Tandem, Bo]leneck, DNN- Derived) Feature Transforma0on (LDA, MLLT, fmllr) GMM HMM N- GRAM Acous&c Model Phone0c inventory Pronuncia0on Lexicon Language Model Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
6 Tandem MLP outputs as input to GMM Input layer Output layer Hidden layers Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
7 Bo]leneck Features Use one narrow hidden layer. Supervised or unsupervised training (autoencoder) Input layer Output layer Hidden layers Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
8 DNN- Derived Features Zhi- Jie Yan, Qiang Huo, Jian Xu: A scalable approach to using DNN- derived features in GMM- HMM based acous0c modeling for LVCSR. INTERSPEECH 2013: Input layer Output layer Hidden layers Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
![DNN- GMM- HMM DNN (Tandem, Bo]leneck, DNN- Derived) Feature](/docs-images/94/120358193/images/9-3.jpg "Transforma0on (LDA, MLLT, fmllr, ) GMM HMM N- GRAM Acous&c Model Phone0c")
9 DNN- GMM- HMM DNN (Tandem, Bo]leneck, DNN- Derived) Feature Transforma0on (LDA, MLLT, fmllr, ) GMM HMM N- GRAM Acous&c Model Phone0c inventory Pronuncia0on Lexicon Language Model Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
10 DNN- HMM DNN HMM N- GRAM Acous&c Model Phone0c inventory Pronuncia0on Lexicon Language Model Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
11 DNN- HMM+RNNLM DNN HMM (N- GRAM +) RNN Acous&c Model Phone0c inventory Pronuncia0on Lexicon Language Model Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
12 RNN- RNNLM RNN Acous0c Model (N- GRAM +) RNN Language Model Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
13 End- to- End RNN Alex Graves et al. (2006) Connec0onist temporal classifica0on: Labelling unsegmented sequence data with recurrent neural networks Proceedings of the Interna0onal Conference on Machine Learning, ICML Florian Eyben, Mar0n Wöllmer, Bjö rn Schuller, Alex Graves (2009) From Speech to Le]ers - Using a Novel Neural Network Architecture for Grapheme Based ASR, ASRU Alex Graves, Navdeep Jaitly, (Jun 2014) Towards End- To- End Speech Recogni0on with Recurrent Neural Networks, Interna0onal Conference on Machine Learning, ICML Jan Chorowski et al. (Dec 2014) End- to- end Con0nuous Speech Recogni0on using A]en0on- based Recurrent NN: First Results, Deep Learning and Representa0on Learning Workshop: NIPS Awni Hannun et al (Dec 2014), Deep Speech: Scaling up end- to- end speech recogni&on, arxiv: [cs.cl] Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
14 End- to- End RNN No perceptual features (MFCC). No feature transforma0on. No phone0c inventory. No transcrip0on dic0onary. No HMM. The output of the RNN are characters including space, apostrophe, (not CD phones) Connec0onist Temporal Classifica0on (No fixed aligment speech/character) Data augmenta0on. 5,000 hours (9600 speakers) + noise = 100,000 hours. Op0miza0ons: data parallelism, model parallelism Good results in noisy condi0ons Adam Coates Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
15 Bidirec0onal Recursive DNN T H E D _ O G Unrolled RNN Spectogram Clipped ReLu Accelerated gradient method GPU friendly h (f) t = g(w (4) h (3) t h (b) t = g(w (4) h (3) t + W r (f) h (f) t 1 + b(4) ) + W r (b) h (b) t+1 + b(4) ) omputed sequentially from to fo Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
16 Connec0onist Temporal Classifica0on T H _ E D O G?"?" How'to'connect'speech'data'with'transcription?' Transcription'not'labeled'per'millisecond' Use'CTC,'from'[Graves'06]' Efficient'dynamic'programming'of'all'possible' alignments'to'compute'error'of'{audio,'transcription}' Bryan'Catanzaro' Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
17 Connec0onist Temporal Classifica0on The framewise network receives an error for misalignment The CTC network predicts the sequence of phonemes / characters (as spikes separated by blanks ) No force alignment (ini0al model) required for training. Waveform label probability dh ax s aw n " " " " dcl d ix v the sound of " " Framewise CTC Alex Graves 2006 Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
18 GMM- HMM / DNN- HMM / RNN EndXtoXend'DL'may'work'better'' when'we'have'big'models'and'' lots'of'data' Accuracy' DL'V1'for'Speech' Traditional'ASR' Deep'Speech' Data'+'Model'Size' Bryan'Catanzaro Adam'Coates' Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
19 Data Augmenta0on This approach needs bigger models and bigger datasets. Synthesis by superposi0on: reverbera0ons, echoes, a large number of short noise recordings from public video sources, ji]er Lombart Effect ' ' >100,000' 80000' Hours' 60000' Synthesized' data' 40000' 20000' 0' 300' 2000' WSJ' Switchboard' Fisher' Deep'Speech' Bryan'Catanzaro Adam'Coates' Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
20 Results 2000 HUB5 (LDC2002S09) System AM training data SWB CH Vesely et al. (2013) SWB Seide et al. (2014) SWB+Fisher+other 13.1 Hannun et al. (2014) SWB+Fisher Zhou et al. (2014) SWB 14.2 Maas et al. (2014) SWB Maas et al. (2014) SWB+Fisher Soltau et al. (2014) SWB Saon et al (2015) SWB+Fisher+CH Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
21 Hybrid(IBM) versus DL(Baidu) Features IBM 2015 Baidu 2014 VTL- PLP, MVN, LDA, STC, fmmlr, i- Vector Alignment GMM- HMM 300K Gaussians - DNN DNN(5x2048) + CNN(128x9x9+5x2048) + +RNN outputs DNN Training HMM CE + MBR Discrimina0ve Training (ST) CTC 32K states (DNN outputs) pentaphone acous0c context Language Model 37M 4- gram + model M (class based exponen0al model) + 2 NNLM 80 log filter banks 4RNN (5 x 2304) 28 outputs - 4- gram (Transcripts) Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
22 Results II Training: Baidu database + data augmenta0on Test: new dataset of 200 recording in both clean and noisy sexngs (no details) Comparison against commercial systems in produc0on Word'Error'Rate'(%)' 50' 40' 30' 20' 10' 0' Clean' Noisy' Combined' Apple'Dictation' Bing'Speech' Google'API' wit.ai' Deep'Speech' Bryan'Catanzaro Adam'Coates' Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
23 Deep Speech Demo h]p:// highlight/
24 References Alex Graves et al. (2006) Connec0onist temporal classifica0on: Labelling unsegmented sequence data with recurrent neural networks Proceedings of the Interna0onal Conference on Machine Learning, ICML Florian Eyben, Mar0n Wöllmer, Bjö rn Schuller, Alex Graves (2009) From Speech to Le]ers - Using a Novel Neural Network Architecture for Grapheme Based ASR, ASRU Alex Graves, Navdeep Jaitly, (Jun 2014) Towards End- To- End Speech Recogni0on with Recurrent Neural Networks, Interna0onal Conference on Machine Learning, ICML Jan Chorowski et al. (Dec 2014) End- to- end Con0nuous Speech Recogni0on using A]en0on- based Recurrent NN: First Results, Deep Learning and Representa0on Learning Workshop: NIPS Awni Hannun et al (Dec 2014), Deep Speech: Scaling up end- to- end speech recogni0on, arxiv: [cs.cl] George Saon et al. The IBM 2015 English Conversa0onal Telephone Speech Recogni0on System, arxiv: [cs.cl] Jose Fonollosa: Deep Learning Architectures for Speech Recogni0on RTTH Summer School, Barcelona, 9- July
Deep Neural Networks in HMM- based and HMM- free Speech RecogniDon
Deep Neural Networks in HMM- based and HMM- free Speech RecogniDon Andrew Maas Collaborators: Awni Hannun, Peng Qi, Chris Lengerich, Ziang Xie, and Anshul Samar Advisors: Andrew Ng and Dan Jurafsky Outline
Lecture 7: Neural network acoustic models in speech recognition
CS 224S / LINGUIST 285 Spoken Language Processing Andrew Maas Stanford University Spring 2017 Lecture 7: Neural network acoustic models in speech recognition Outline Hybrid acoustic modeling overview Basic
arxiv: v1 [cs.cl] 28 Nov 2016
![arxiv: v1 [cs.cl] 28 Nov 2016](/thumbs/80/82342129.jpg "arxiv: v1 [cs.cl] 28 Nov 2016")
An End-to-End Architecture for Keyword Spotting and Voice Activity Detection arxiv:1611.09405v1 [cs.cl] 28 Nov 2016 Chris Lengerich Mindori Palo Alto, CA chris@mindori.com Abstract Awni Hannun Mindori
LOW-RANK MATRIX FACTORIZATION FOR DEEP NEURAL NETWORK TRAINING WITH HIGH-DIMENSIONAL OUTPUT TARGETS
LOW-RANK MATRIX FACTORIZATION FOR DEEP NEURAL NETWORK TRAINING WITH HIGH-DIMENSIONAL OUTPUT TARGETS Tara N. Sainath, Brian Kingsbury, Vikas Sindhwani, Ebru Arisoy, Bhuvana Ramabhadran IBM T. J. Watson
Speech Recognition with Quaternion Neural Networks
Speech Recognition with Quaternion Neural Networks LIA - 2019 Titouan Parcollet, Mohamed Morchid, Georges Linarès University of Avignon, France ORKIS, France Summary I. Problem definition II. Quaternion
Scaling Deep Learning. Bryan
Scaling Deep Learning @ctnzr What do we want AI to do? Guide us to content Keep us organized Help us find things Help us communicate 帮助我们沟通 Drive us to work Serve drinks? Image Q&A Baidu IDL Sample questions
Why DNN Works for Speech and How to Make it More Efficient?
Why DNN Works for Speech and How to Make it More Efficient? Hui Jiang Department of Electrical Engineering and Computer Science Lassonde School of Engineering, York University, CANADA Joint work with Y.
Long Short-Term Memory Recurrent Neural Network Architectures for Large Scale Acoustic Modeling
INTERSPEECH 2014 Long Short-Term Memory Recurrent Neural Network Architectures for Large Scale Acoustic Modeling Haşim Sak, Andrew Senior, Françoise Beaufays Google, USA {hasim,andrewsenior,fsb@google.com}
Gated Recurrent Unit Based Acoustic Modeling with Future Context
Interspeech 2018 2-6 September 2018, Hyderabad Gated Recurrent Unit Based Acoustic Modeling with Future Context Jie Li 1, Xiaorui Wang 1, Yuanyuan Zhao 2, Yan Li 1 1 Kwai, Beijing, P.R. China 2 Institute
END-TO-END CHINESE TEXT RECOGNITION
END-TO-END CHINESE TEXT RECOGNITION Jie Hu 1, Tszhang Guo 1, Ji Cao 2, Changshui Zhang 1 1 Department of Automation, Tsinghua University 2 Beijing SinoVoice Technology November 15, 2017 Presentation at
A Comparison of Sequence-Trained Deep Neural Networks and Recurrent Neural Networks Optical Modeling For Handwriting Recognition
A Comparison of Sequence-Trained Deep Neural Networks and Recurrent Neural Networks Optical Modeling For Handwriting Recognition Théodore Bluche, Hermann Ney, Christopher Kermorvant SLSP 14, Grenoble October
Making Deep Belief Networks Effective for Large Vocabulary Continuous Speech Recognition
Making Deep Belief Networks Effective for Large Vocabulary Continuous Speech Recognition Tara N. Sainath 1, Brian Kingsbury 1, Bhuvana Ramabhadran 1, Petr Fousek 2, Petr Novak 2, Abdel-rahman Mohamed 3
Adversarial Feature-Mapping for Speech Enhancement
Interspeech 2018 2-6 September 2018, Hyderabad Adversarial Feature-Mapping for Speech Enhancement Zhong Meng 1,2, Jinyu Li 1, Yifan Gong 1, Biing-Hwang (Fred) Juang 2 1 Microsoft AI and Research, Redmond,
End-to-End Training of Acoustic Models for Large Vocabulary Continuous Speech Recognition with TensorFlow
End-to-End Training of Acoustic Models for Large Vocabulary Continuous Speech Recognition with TensorFlow Ehsan Variani, Tom Bagby, Erik McDermott, Michiel Bacchiani Google Inc, Mountain View, CA, USA
Improving CTC-based Acoustic Model with Very Deep Residual Time-delay Neural Networks
Interspeech 2018 2-6 September 2018, Hyderabad Improving CTC-based Acoustic Model with Very Deep Residual Time-delay Neural Networks Sheng Li 1, Xugang Lu 1, Ryoichi Takashima 1, Peng Shen 1, Tatsuya Kawahara
SYNTHESIZED STEREO MAPPING VIA DEEP NEURAL NETWORKS FOR NOISY SPEECH RECOGNITION
2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) SYNTHESIZED STEREO MAPPING VIA DEEP NEURAL NETWORKS FOR NOISY SPEECH RECOGNITION Jun Du 1, Li-Rong Dai 1, Qiang Huo
Speech Recogni,on using HTK CS4706. Fadi Biadsy April 21 st, 2008
peech Recogni,on using HTK C4706 Fadi Biadsy April 21 st, 2008 1 Outline peech Recogni,on Feature Extrac,on HMM 3 basic problems HTK teps to Build a speech recognizer 2 peech Recogni,on peech ignal to
Joint Optimisation of Tandem Systems using Gaussian Mixture Density Neural Network Discriminative Sequence Training
Joint Optimisation of Tandem Systems using Gaussian Mixture Density Neural Network Discriminative Sequence Training Chao Zhang and Phil Woodland March 8, 07 Cambridge University Engineering Department
GPU Accelerated Model Combination for Robust Speech Recognition and Keyword Search
GPU Accelerated Model Combination for Robust Speech Recognition and Keyword Search Wonkyum Lee Jungsuk Kim Ian Lane Electrical and Computer Engineering Carnegie Mellon University March 26, 2014 @GTC2014
SPEECH FEATURE DENOISING AND DEREVERBERATION VIA DEEP AUTOENCODERS FOR NOISY REVERBERANT SPEECH RECOGNITION. Xue Feng, Yaodong Zhang, James Glass
2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) SPEECH FEATURE DENOISING AND DEREVERBERATION VIA DEEP AUTOENCODERS FOR NOISY REVERBERANT SPEECH RECOGNITION Xue Feng,
Hello Edge: Keyword Spotting on Microcontrollers
Hello Edge: Keyword Spotting on Microcontrollers Yundong Zhang, Naveen Suda, Liangzhen Lai and Vikas Chandra ARM Research, Stanford University arxiv.org, 2017 Presented by Mohammad Mofrad University of
ACOUSTIC MODELING WITH NEURAL GRAPH EMBEDDINGS. Yuzong Liu, Katrin Kirchhoff
ACOUSTIC MODELING WITH NEURAL GRAPH EMBEDDINGS Yuzong Liu, Katrin Kirchhoff Department of Electrical Engineering University of Washington, Seattle, WA 98195 ABSTRACT Graph-based learning (GBL) is a form
GMM-FREE DNN TRAINING. Andrew Senior, Georg Heigold, Michiel Bacchiani, Hank Liao
GMM-FREE DNN TRAINING Andrew Senior, Georg Heigold, Michiel Bacchiani, Hank Liao Google Inc., New York {andrewsenior,heigold,michiel,hankliao}@google.com ABSTRACT While deep neural networks (DNNs) have
Discriminative training and Feature combination
Discriminative training and Feature combination Steve Renals Automatic Speech Recognition ASR Lecture 13 16 March 2009 Steve Renals Discriminative training and Feature combination 1 Overview Hot topics
CUED-RNNLM An Open-Source Toolkit for Efficient Training and Evaluation of Recurrent Neural Network Language Models
CUED-RNNLM An Open-Source Toolkit for Efficient Training and Evaluation of Recurrent Neural Network Language Models Xie Chen, Xunying Liu, Yanmin Qian, Mark Gales and Phil Woodland April 1, 2016 Overview
The Hitachi/JHU CHiME-5 system: Advances in speech recognition for everyday home environments using multiple microphone arrays
CHiME2018 workshop The Hitachi/JHU CHiME-5 system: Advances in speech recognition for everyday home environments using multiple microphone arrays Naoyuki Kanda 1, Rintaro Ikeshita 1, Shota Horiguchi 1,
Variable-Component Deep Neural Network for Robust Speech Recognition
Variable-Component Deep Neural Network for Robust Speech Recognition Rui Zhao 1, Jinyu Li 2, and Yifan Gong 2 1 Microsoft Search Technology Center Asia, Beijing, China 2 Microsoft Corporation, One Microsoft
Feature Engineering in Context-Dependent Deep Neural Networks for Conversational Speech Transcription
Feature Engineering in Context-Dependent Deep Neural Networks for Conversational Speech Transcription Frank Seide 1,GangLi 1, Xie Chen 1,2, and Dong Yu 3 1 Microsoft Research Asia, 5 Danling Street, Haidian
10703 Deep Reinforcement Learning and Control
10703 Deep Reinforcement Learning and Control Russ Salakhutdinov Machine Learning Department rsalakhu@cs.cmu.edu Policy Gradient II Used Materials Disclaimer: Much of the material and slides for this lecture
Low latency acoustic modeling using temporal convolution and LSTMs
1 Low latency acoustic modeling using temporal convolution and LSTMs Vijayaditya Peddinti, Yiming Wang, Daniel Povey, Sanjeev Khudanpur Abstract Bidirectional long short term memory (BLSTM) acoustic models
Improving Bottleneck Features for Automatic Speech Recognition using Gammatone-based Cochleagram and Sparsity Regularization
Improving Bottleneck Features for Automatic Speech Recognition using Gammatone-based Cochleagram and Sparsity Regularization Chao Ma 1,2,3, Jun Qi 4, Dongmei Li 1,2,3, Runsheng Liu 1,2,3 1. Department
A new approach for supervised power disaggregation by using a deep recurrent LSTM network
A new approach for supervised power disaggregation by using a deep recurrent LSTM network GlobalSIP 2015, 14th Dec. Lukas Mauch and Bin Yang Institute of Signal Processing and System Theory University
An Online Sequence-to-Sequence Model Using Partial Conditioning
An Online Sequence-to-Sequence Model Using Partial Conditioning Navdeep Jaitly Google Brain ndjaitly@google.com David Sussillo Google Brain sussillo@google.com Quoc V. Le Google Brain qvl@google.com Oriol
Bidirectional Truncated Recurrent Neural Networks for Efficient Speech Denoising
Bidirectional Truncated Recurrent Neural Networks for Efficient Speech Denoising Philémon Brakel, Dirk Stroobandt, Benjamin Schrauwen Department of Electronics and Information Systems, Ghent University,
Bo#leneck Features from SNR- Adap9ve Denoising Deep Classifier for Speaker Iden9fica9on
Bo#leneck Features from SNR- Adap9ve Denoising Deep Classifier for Speaker Iden9fica9on TAN Zhili & MAK Man-Wai APSIPA 2015 Department of Electronic and Informa2on Engineering The Hong Kong Polytechnic
Introducing CURRENNT: The Munich Open-Source CUDA RecurREnt Neural Network Toolkit
Journal of Machine Learning Research 6 205) 547-55 Submitted 7/3; Published 3/5 Introducing CURRENNT: The Munich Open-Source CUDA RecurREnt Neural Network Toolkit Felix Weninger weninger@tum.de Johannes
Training LDCRF model on unsegmented sequences using Connectionist Temporal Classification
Training LDCRF model on unsegmented sequences using Connectionist Temporal Classification 1 Amir Ahooye Atashin, 2 Kamaledin Ghiasi-Shirazi, 3 Ahad Harati Department of Computer Engineering Ferdowsi University
Learning and Modeling Unit Embeddings for Improving HMM-based Unit Selection Speech Synthesis
Interspeech 2018 2-6 September 2018, Hyderabad Learning and Modeling Unit Embeddings for Improving HMM-based Unit Selection Synthesis Xiao Zhou, Zhen-Hua Ling, Zhi-Ping Zhou, Li-Rong Dai National Engineering
Voice command module for Smart Home Automation
Voice command module for Smart Home Automation LUKA KRALJEVIĆ, MLADEN RUSSO, MAJA STELLA Laboratory for Smart Environment Technologies, University of Split, FESB Ruđera Boškovića 32, 21000, Split CROATIA
A Deep Learning primer
A Deep Learning primer Riccardo Zanella r.zanella@cineca.it SuperComputing Applications and Innovation Department 1/21 Table of Contents Deep Learning: a review Representation Learning methods DL Applications
Sequence Modeling: Recurrent and Recursive Nets. By Pyry Takala 14 Oct 2015
Sequence Modeling: Recurrent and Recursive Nets By Pyry Takala 14 Oct 2015 Agenda Why Recurrent neural networks? Anatomy and basic training of an RNN (10.2, 10.2.1) Properties of RNNs (10.2.2, 8.2.6) Using
arxiv: v1 [cs.cv] 4 Feb 2018
![arxiv: v1 [cs.cv] 4 Feb 2018](/thumbs/90/103254317.jpg "arxiv: v1 [cs.cv] 4 Feb 2018")
End2You The Imperial Toolkit for Multimodal Profiling by End-to-End Learning arxiv:1802.01115v1 [cs.cv] 4 Feb 2018 Panagiotis Tzirakis Stefanos Zafeiriou Björn W. Schuller Department of Computing Imperial
Code Mania Artificial Intelligence: a. Module - 1: Introduction to Artificial intelligence and Python:
Code Mania 2019 Artificial Intelligence: a. Module - 1: Introduction to Artificial intelligence and Python: 1. Introduction to Artificial Intelligence 2. Introduction to python programming and Environment
27: Hybrid Graphical Models and Neural Networks
10-708: Probabilistic Graphical Models 10-708 Spring 2016 27: Hybrid Graphical Models and Neural Networks Lecturer: Matt Gormley Scribes: Jakob Bauer Otilia Stretcu Rohan Varma 1 Motivation We first look
arxiv: v2 [cs.lg] 7 Jan 2016
![arxiv: v2 [cs.lg] 7 Jan 2016](/thumbs/80/80585793.jpg "arxiv: v2 [cs.lg] 7 Jan 2016")
RECURRENT MODELS FOR AUDITORY ATTENTION IN MULTI-MICROPHONE DISTANCE SPEECH RECOGNITION Suyoun Kim Electrical Computer Engineering Carnegie Mellon University suyoun@cmu.edu Ian Lane Electrical Computer
Convolutional Sequence to Sequence Non-intrusive Load Monitoring
1 Convolutional Sequence to Sequence Non-intrusive Load Monitoring Kunjin Chen, Qin Wang, Ziyu He Kunlong Chen, Jun Hu and Jinliang He Department of Electrical Engineering, Tsinghua University, Beijing,
Speech Technology Using in Wechat
Speech Technology Using in Wechat FENG RAO Powered by WeChat Outline Introduce Algorithm of Speech Recognition Acoustic Model Language Model Decoder Speech Technology Open Platform Framework of Speech
Factored deep convolutional neural networks for noise robust speech recognition
INTERSPEECH 2017 August 20 24, 2017, Stockholm, Sweden Factored deep convolutional neural networks for noise robust speech recognition Masakiyo Fujimoto National Institute of Information and Communications
Query-by-example spoken term detection based on phonetic posteriorgram Query-by-example spoken term detection based on phonetic posteriorgram
International Conference on Education, Management and Computing Technology (ICEMCT 2015) Query-by-example spoken term detection based on phonetic posteriorgram Query-by-example spoken term detection based
Handwritten Text Recognition
Handwritten Text Recognition M.J. Castro-Bleda, S. España-Boquera, F. Zamora-Martínez Universidad Politécnica de Valencia Spain Avignon, 9 December 2010 Text recognition () Avignon Avignon, 9 December
Pair-wise Distance Metric Learning of Neural Network Model for Spoken Language Identification
INTERSPEECH 2016 September 8 12, 2016, San Francisco, USA Pair-wise Distance Metric Learning of Neural Network Model for Spoken Language Identification 2 1 Xugang Lu 1, Peng Shen 1, Yu Tsao 2, Hisashi
A PRIORITIZED GRID LONG SHORT-TERM MEMORY RNN FOR SPEECH RECOGNITION. Wei-Ning Hsu, Yu Zhang, and James Glass
A PRIORITIZED GRID LONG SHORT-TERM MEMORY RNN FOR SPEECH RECOGNITION Wei-Ning Hsu, Yu Zhang, and James Glass Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology
Recurrent Neural Networks
Recurrent Neural Networks 11-785 / Fall 2018 / Recitation 7 Raphaël Olivier Recap : RNNs are magic They have infinite memory They handle all kinds of series They re the basis of recent NLP : Translation,
저작권법에따른이용자의권리는위의내용에의하여영향을받지않습니다.
저작자표시 - 비영리 - 변경금지 2.0 대한민국 이용자는아래의조건을따르는경우에한하여자유롭게 이저작물을복제, 배포, 전송, 전시, 공연및방송할수있습니다. 다음과같은조건을따라야합니다 : 저작자표시. 귀하는원저작자를표시하여야합니다. 비영리. 귀하는이저작물을영리목적으로이용할수없습니다. 변경금지. 귀하는이저작물을개작, 변형또는가공할수없습니다. 귀하는, 이저작물의재이용이나배포의경우,
Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank text
Philosophische Fakultät Seminar für Sprachwissenschaft Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank text 06 July 2017, Patricia Fischer & Neele Witte Overview Sentiment
Rolling Bearing Diagnosis Based on CNN-LSTM and Various Condition Dataset
Rolling Bearing Diagnosis Based on CNN-LSTM and Various Condition Dataset Osamu Yoshimatsu 1, Yoshihiro Satou 2, and Kenichi Shibasaki 3 1,2,3 Core Technology R&D Center, NSK Ltd., Fujisawa, Kanagawa,
Pixel-level Generative Model
Pixel-level Generative Model Generative Image Modeling Using Spatial LSTMs (2015NIPS) L. Theis and M. Bethge University of Tübingen, Germany Pixel Recurrent Neural Networks (2016ICML) A. van den Oord,
Deep Learning on Arm Cortex-M Microcontrollers. Rod Crawford Director Software Technologies, Arm
Deep Learning on Arm Cortex-M Microcontrollers Rod Crawford Director Software Technologies, Arm What is Machine Learning (ML)? Artificial Intelligence Machine Learning Deep Learning Neural Networks Additional
Practical Methodology. Lecture slides for Chapter 11 of Deep Learning Ian Goodfellow
Practical Methodology Lecture slides for Chapter 11 of Deep Learning www.deeplearningbook.org Ian Goodfellow 2016-09-26 What drives success in ML? Arcane knowledge of dozens of obscure algorithms? Mountains
RECURRENT NEURAL NETWORKS
RECURRENT NEURAL NETWORKS Methods Traditional Deep-Learning based Non-machine Learning Machine-Learning based method Supervised SVM MLP CNN RNN (LSTM) Localizati on GPS, SLAM Self Driving Perception Pedestrian
SVD-based Universal DNN Modeling for Multiple Scenarios
SVD-based Universal DNN Modeling for Multiple Scenarios Changliang Liu 1, Jinyu Li 2, Yifan Gong 2 1 Microsoft Search echnology Center Asia, Beijing, China 2 Microsoft Corporation, One Microsoft Way, Redmond,
Restricted Boltzmann Machines. Shallow vs. deep networks. Stacked RBMs. Boltzmann Machine learning: Unsupervised version
Shallow vs. deep networks Restricted Boltzmann Machines Shallow: one hidden layer Features can be learned more-or-less independently Arbitrary function approximator (with enough hidden units) Deep: two
How to Build Optimized ML Applications with Arm Software
How to Build Optimized ML Applications with Arm Software Arm Technical Symposia 2018 ML Group Overview Today we will talk about applied machine learning (ML) on Arm. My aim for today is to show you just
A comprehensive study of hybrid neural network hidden Markov model for offline handwritten Chinese text recognition
International Journal on Document Analysis and Recognition (IJDAR) https://doi.org/10.1007/s10032-018-0307-0 ORIGINAL PAPER A comprehensive study of hybrid neural network hidden Markov model for offline
MIXED PRECISION TRAINING: THEORY AND PRACTICE Paulius Micikevicius
MIXED PRECISION TRAINING: THEORY AND PRACTICE Paulius Micikevicius What is Mixed Precision Training? Reduced precision tensor math with FP32 accumulation, FP16 storage Successfully used to train a variety
Machine Learning: Overview & Applications to Test
Machine Learning: Overview & Applications to Test 1st Lt Takayuki Iguchi 1st Lt Megan E. Lewis AFOTEC/Det 5/DTS Release Date: 6 MAR 17 Approved for Public Release: Distribution Unlimited. AFOTEC Public
Recurrent Neural Network (RNN) Industrial AI Lab.
Recurrent Neural Network (RNN) Industrial AI Lab. For example (Deterministic) Time Series Data Closed- form Linear difference equation (LDE) and initial condition High order LDEs 2 (Stochastic) Time Series
Unsupervised Feature Learning for Optical Character Recognition
Unsupervised Feature Learning for Optical Character Recognition Devendra K Sahu and C. V. Jawahar Center for Visual Information Technology, IIIT Hyderabad, India. Abstract Most of the popular optical character
Deep Neural Networks Applications in Handwriting Recognition
Deep Neural Networks Applications in Handwriting Recognition 2 Who am I? Théodore Bluche PhD defended at Université Paris-Sud last year Deep Neural Networks for Large Vocabulary Handwritten
Deep Learning. Yee Whye Teh (Oxford Statistics & DeepMind)
Deep Learning Yee Whye Teh (Oxford Statistics & DeepMind) http://csml.stats.ox.ac.uk/people/teh What is Machine Learning? Information Structure Prediction Decisions Actions data What is Machine Learning?
arxiv: v1 [cs.cl] 30 Jan 2018
![arxiv: v1 [cs.cl] 30 Jan 2018](/thumbs/96/127233064.jpg "arxiv: v1 [cs.cl] 30 Jan 2018")
ACCELERATING RECURRENT NEURAL NETWORK LANGUAGE MODEL BASED ONLINE SPEECH RECOGNITION SYSTEM Kyungmin Lee, Chiyoun Park, Namhoon Kim, and Jaewon Lee DMC R&D Center, Samsung Electronics, Seoul, Korea {k.m.lee,
Deep Neural Networks Applications in Handwriting Recognition
Deep Neural Networks Applications in Handwriting Recognition Théodore Bluche theodore.bluche@gmail.com São Paulo Meetup - 9 Mar. 2017 2 Who am I? Théodore Bluche PhD defended
Frame and Segment Level Recurrent Neural Networks for Phone Classification
Frame and Segment Level Recurrent Neural Networks for Phone Classification Martin Ratajczak 1, Sebastian Tschiatschek 2, Franz Pernkopf 1 1 Graz University of Technology, Signal Processing and Speech Communication
Binary Deep Neural Networks for Speech Recognition
INTERSPEECH 2017 August 20 24, 2017, Stockholm, Sweden Binary Deep Neural Networks for Speech Recognition Xu Xiang, Yanmin Qian, Kai Yu Key Lab. of Shanghai Education Commission for Intelligent Interaction
Manifold Constrained Deep Neural Networks for ASR
1 Manifold Constrained Deep Neural Networks for ASR Department of Electrical and Computer Engineering, McGill University Richard Rose and Vikrant Tomar Motivation Speech features can be characterized as
ABSTRACT 1. INTRODUCTION
LOW-RANK PLUS DIAGONAL ADAPTATION FOR DEEP NEURAL NETWORKS Yong Zhao, Jinyu Li, and Yifan Gong Microsoft Corporation, One Microsoft Way, Redmond, WA 98052, USA {yonzhao; jinyli; ygong}@microsoft.com ABSTRACT
CSC 578 Neural Networks and Deep Learning
CSC 578 Neural Networks and Deep Learning Fall 2018/19 7. Recurrent Neural Networks (Some figures adapted from NNDL book) 1 Recurrent Neural Networks 1. Recurrent Neural Networks (RNNs) 2. RNN Training
Convolutional Sequence to Sequence Learning. Denis Yarats with Jonas Gehring, Michael Auli, David Grangier, Yann Dauphin Facebook AI Research
Convolutional Sequence to Sequence Learning Denis Yarats with Jonas Gehring, Michael Auli, David Grangier, Yann Dauphin Facebook AI Research Sequence generation Need to model a conditional distribution
Slide credit from Hung-Yi Lee & Richard Socher
Slide credit from Hung-Yi Lee & Richard Socher 1 Review Word Vector 2 Word2Vec Variants Skip-gram: predicting surrounding words given the target word (Mikolov+, 2013) CBOW (continuous bag-of-words): predicting
Deep Learning Applications
October 20, 2017 Overview Supervised Learning Feedforward neural network Convolution neural network Recurrent neural network Recursive neural network (Recursive neural tensor network) Unsupervised Learning
Lecture 13. Deep Belief Networks. Michael Picheny, Bhuvana Ramabhadran, Stanley F. Chen, Markus Nussbaum-Thom
Lecture 13 Deep Belief Networks Michael Picheny, Bhuvana Ramabhadran, Stanley F. Chen, Markus Nussbaum-Thom Watson Group IBM T.J. Watson Research Center Yorktown Heights, New York, USA {picheny,bhuvana,stanchen,nussbaum}@us.ibm.com
Segmentation-free Vehicle License Plate Recognition using ConvNet-RNN
Segmentation-free Vehicle License Plate Recognition using ConvNet-RNN Teik Koon Cheang (Author) cheangtk@1utar.my Yong Shean Chong yshean@1utar.my Yong Haur Tay tayyh@utar.edu.my Abstract While vehicle
Lecture 8. LVCSR Training and Decoding. Michael Picheny, Bhuvana Ramabhadran, Stanley F. Chen
Lecture 8 LVCSR Training and Decoding Michael Picheny, Bhuvana Ramabhadran, Stanley F. Chen IBM T.J. Watson Research Center Yorktown Heights, New York, USA {picheny,bhuvana,stanchen}@us.ibm.com 12 November
Machine Learning for Speaker Recogni2on and Bioinforma2cs
Machine Learning for Speaker Recogni2on and Bioinforma2cs Man-Wai MAK Dept of Electronic and Informa8on Engineering, The Hong Kong Polytechnic University http://wwweiepolyueduhk/~mwmak/ UTS/PolyU Workshop
OPENSEQ2SEQ: A DEEP LEARNING TOOLKIT FOR SPEECH RECOGNITION, SPEECH SYNTHESIS, AND NLP
OPENSEQ2SEQ: A DEEP LEARNING TOOLKIT FOR SPEECH RECOGNITION, SPEECH SYNTHESIS, AND NLP Boris Ginsburg, Vitaly Lavrukhin, Igor Gitman, Oleksii Kuchaiev, Jason Li, Vahid Noroozi, Ravi Teja Gadde, Chip Nguyen
BUILDING CORPORA OF TRANSCRIBED SPEECH FROM OPEN ACCESS SOURCES
BUILDING CORPORA OF TRANSCRIBED SPEECH FROM OPEN ACCESS SOURCES O.O. Iakushkin a, G.A. Fedoseev, A.S. Shaleva, O.S. Sedova Saint Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg,
Hybrid Accelerated Optimization for Speech Recognition
INTERSPEECH 26 September 8 2, 26, San Francisco, USA Hybrid Accelerated Optimization for Speech Recognition Jen-Tzung Chien, Pei-Wen Huang, Tan Lee 2 Department of Electrical and Computer Engineering,
LSTM: An Image Classification Model Based on Fashion-MNIST Dataset
LSTM: An Image Classification Model Based on Fashion-MNIST Dataset Kexin Zhang, Research School of Computer Science, Australian National University Kexin Zhang, U6342657@anu.edu.au Abstract. The application
Multimodal Gesture Recognition using Multi-stream Recurrent Neural Network
Multimodal Gesture Recognition using Multi-stream Recurrent Neural Network Noriki Nishida and Hideki Nakayama Machine Perception Group Graduate School of Information Science and Technology The University
Does speech enhancement work with end-to-end ASR objectives?: Experimental analysis of multichannel end-to-end ASR
MITSUBISHI ELECTRIC RESEARCH LABORATORIES http://www.merl.com Does speech enhancement work with end-to-end ASR objectives?: Experimental analysis of multichannel end-to-end ASR Ochiai, T.; Watanabe, S.;
Machine Learning. Deep Learning. Eric Xing (and Pengtao Xie) , Fall Lecture 8, October 6, Eric CMU,
Machine Learning 10-701, Fall 2015 Deep Learning Eric Xing (and Pengtao Xie) Lecture 8, October 6, 2015 Eric Xing @ CMU, 2015 1 A perennial challenge in computer vision: feature engineering SIFT Spin image
DEEP LEARNING REVIEW. Yann LeCun, Yoshua Bengio & Geoffrey Hinton Nature Presented by Divya Chitimalla
DEEP LEARNING REVIEW Yann LeCun, Yoshua Bengio & Geoffrey Hinton Nature 2015 -Presented by Divya Chitimalla What is deep learning Deep learning allows computational models that are composed of multiple
SPEECH FEATURE EXTRACTION USING WEIGHTED HIGHER-ORDER LOCAL AUTO-CORRELATION
Far East Journal of Electronics and Communications Volume 3, Number 2, 2009, Pages 125-140 Published Online: September 14, 2009 This paper is available online at http://www.pphmj.com 2009 Pushpa Publishing
Encoding RNNs, 48 End of sentence (EOS) token, 207 Exploding gradient, 131 Exponential function, 42 Exponential Linear Unit (ELU), 44
A Activation potential, 40 Annotated corpus add padding, 162 check versions, 158 create checkpoints, 164, 166 create input, 160 create train and validation datasets, 163 dropout, 163 DRUG-AE.rel file,
Learning The Lexicon!
Learning The Lexicon! A Pronunciation Mixture Model! Ian McGraw! (imcgraw@mit.edu)! Ibrahim Badr Jim Glass! Computer Science and Artificial Intelligence Lab! Massachusetts Institute of Technology! Cambridge,
Compressing Deep Neural Networks using a Rank-Constrained Topology
Compressing Deep Neural Networks using a Rank-Constrained Topology Preetum Nakkiran Department of EECS University of California, Berkeley, USA preetum@berkeley.edu Raziel Alvarez, Rohit Prabhavalkar, Carolina
arxiv: v5 [cs.lg] 21 Feb 2014
![arxiv: v5 [cs.lg] 21 Feb 2014](/thumbs/89/100981620.jpg "arxiv: v5 [cs.lg] 21 Feb 2014")
Do Deep Nets Really Need to be Deep? arxiv:1312.6184v5 [cs.lg] 21 Feb 2014 Lei Jimmy Ba University of Toronto jimmy@psi.utoronto.ca Abstract Rich Caruana Microsoft Research rcaruana@microsoft.com Currently,
arxiv: v1 [cs.lg] 20 May 2018
![arxiv: v1 [cs.lg] 20 May 2018](/thumbs/90/102436424.jpg "arxiv: v1 [cs.lg] 20 May 2018")
Targeted Adversarial Examples for Black Box Audio Systems Rohan Taori Amog Kamsetty Brenton Chu Nikita Vemuri arxiv:1805.07820v1 [cs.lg] 20 May 2018 University of California, Berkeley Machine Learning
All You Want To Know About CNNs. Yukun Zhu
All You Want To Know About CNNs Yukun Zhu Deep Learning Deep Learning Image from http://imgur.com/ Deep Learning Image from http://imgur.com/ Deep Learning Image from http://imgur.com/ Deep Learning Image
Deep Learning for Vision: Tricks of the Trade
Deep Learning for Vision: Tricks of the Trade Marc'Aurelio Ranzato Facebook, AI Group www.cs.toronto.edu/~ranzato BAVM Friday, 4 October 2013 Ideal Features Ideal Feature Extractor - window, right - chair,