Statistical Machine Translation. Machine Translation Phrase-Based Statistical MT. Motivation for Phrase-based SMT
|
|
- Ambrose Haynes
- 5 years ago
- Views:
Transcription
1 Statistical Machine Translation Machine Translation Phrase-Based Statistical MT Jörg Tiedemann Department of Linguistics and Philology Uppsala University October 2009 Probabilistic view on MT (E = target language, F = source language): Ê = argmax E P(E F ) = argmax E P(F E)P(E) Jörg Tiedemann 1/1 Jörg Tiedemann 2/1 Statistical Machine Translation: Language Modeling Motivation for Phrase-based SMT Language modeling: (probabilistic) LM = predict likelihood of a given string What is the likelihood P(E) to observe sentence E? Exactly what we need! Estimate probabilities from corpora: decompose into N-grams! unigram model: P(E) = P(e 1 ) P(e 2 )...P(e n ) bigram model: P(E) = P(e 1 ) P(e 2 e 1 ) P(e 3 e 2 )...P(e n e n 1 ) trigram model: P(E) = P(e 1 ) P(e 2 e 1 ) P(e 3 e 1, e 2 )...P(e n e n 2 e n 1, ) There would be much more to say about language modeling... Word-based SMT statistical word alignment P(F E) language modeling P(E) global decoding argmax E P(F E)P(E) Word-by-word translation is too weak! contextual dependencies non-compositional constructions n:m relations look at larger chunks! Jörg Tiedemann 3/1 Jörg Tiedemann 4/1
2 Phrase-based SMT Phrase-based SMT Translation model in PSMT: Motivation phrases = word N-grams less ambiguity, more context in translation table handle non-compositional expressions local reorderings covered by phrase translations distortion : reordering on phrase level P(F E) = I φ(f i e i )d(start i, end i 1 ) i=1 phrases are extracted from word aligned parallel corpora phrase translation probabilities (MLE): φ(f e) = count(f, e) count(f, e) f Moses toolkit: ( Jörg Tiedemann 5/1 Jörg Tiedemann 6/1 Phrase-based SMT Statistical word alignment Standard models: Phrase translation probabilities: need phrase alignments in parallel corpus induce them from word alignments (IBM models) score extracted phrases (MLE) IBM models 1-5 (cascaded), EM training, final parameters: word translation probabilities (lexical model) fertility probabilities distortion probabilities (reordering) Viterbi alignment assign most likely links between words according to the statistical word alignment model from above Jörg Tiedemann 7/1 Jörg Tiedemann 8/1
3 Viterbi Word Alignment Viterbi Word Alignment from GIZA++ From the German-English Europarl corpus: special NULL word (NULL la) EMPTY alignment possible (did) only 1:many (slap); not many:1 depending on alignment direction Alignment tool: GIZA++ ( # Sentence pair (5) source length 12 target length 11 alignment score : e-24 ich bitte sie, sich zu einer schweigeminute zu erheben. NULL ({ }) please ({ }) rise ({ }), ({ 4 }) then ({ 5 }), ({ }) for ({ 6 }) this ({ 7 }) minute ({ 8 }) ({ }) s ({ }) silence ({ 9 10 }). ({ 11 }) # Sentence pair (6) source length 12 target length 10 alignment score : e-15 ( das parlament erhebt sich zu einer schweigeminute. ) NULL ({ }) ( ({ 1 }) the ({ 2 }) house ({ 3 }) rose ({ 4 5 }) and ({ }) observed ({ 6 }) a ({ 7 }) minute ({ 8 }) ({ }) s ({ }) silence ({ 9 }) ) ({ 10 }) Jörg Tiedemann 9/1 Jörg Tiedemann 11/1 Viterbi Word Alignment Word Alignment Symmetrization source-to-target word alignment: Asymmetric alignment! no n:1 alignments can run IBM models in both directions! different links in source-to-target and target-to-source best alignment = merge both directions (?!) How? Symmetrization heuristics! Jörg Tiedemann 12/1 Jörg Tiedemann 13/1
4 Word Alignment Symmetrization target-to-source word alignment: Word Alignment Symmetrization symmetrized word alignment: Jörg Tiedemann 14/1 Jörg Tiedemann 15/1 Word Alignment Symmetrization start with intersection, add adjacent links (from union)... Phrase extraction Get ALL phrase pairs that are consistent with word alignments Jörg Tiedemann 16/1 Jörg Tiedemann 17/1
5 Phrase extraction Phrase extraction Jörg Tiedemann 18/1 Jörg Tiedemann 19/1 Phrase extraction Phrase extraction Jörg Tiedemann 20/1 Jörg Tiedemann 21/1
6 Phrase extraction Phrase extraction Jörg Tiedemann 22/1 Jörg Tiedemann 23/1 Scoring phrases Phrase tables Examples from a phrase table (Pirates of the Caribbean): Simple Maximum likelihood estimation: φ(f e) = count(f, e) count(f, e) f A huge phrase table! (with a lot of garbage?) Swedish English Score, det?, it s , det?, that s 1 att bli besvikna be disappointed 1 att bli en sj?v to becoming one 1 bara vi just 0.1 bara just 0.6 bara only barbossa och hans bes?tning barbossa and his crew 1 barbossa och hans barbossa and his 1 barbossa t?ker g?a. allt barbossa is up to all 1 (The training set was too small to get reasonable counts!) Jörg Tiedemann 24/1 Jörg Tiedemann 25/1
7 The final model for PB-SMT PB-SMT extension: Log-linear Models Instead of noisy-channel model Ê = argmax EP(F E)P(E): Ê = argmax E P(E F ) = argmax E ( φ(f i e i ) d(start i, end i 1 ) P(E) ω length(e)) model posterior directly: Ê = argmax EP(E F ) many feature functions h m (E, F ) may influence P(E F) Distortion d: Chance to move phrases to other positions fixed distortion limit (e.g. 6) simple penalty for moving: α start i end i 1 1 OR lexicalized distortion (learned from alignment) phrase translation model E F phrase translation model F E lexical weights from underlying word alignment a language model P(E) lexicalized reordering model length features (word/phrase costs/penalties) Word cost: ω length(e) = bias for longer output P(E F) = weighted combination of feature functions! Jörg Tiedemann 26/1 Jörg Tiedemann 27/1 PB-SMT extension: Log-linear Models P(E F ) = weighted (λ m ) combination of feature functions (h m ) P(E F ) = ep M m=1 λ mh m (E,F) Z Ê = argmax E P(E F) = argmax E (logp(e F )) How to learn weights λ m? M = argmax E λ m h m (E, F ) m=1 Minimum error rate training (MERT) on development set! Measure error in terms of BLEU scores (n-best list) Iterative adjustment of model parameters (slow but effective!) Phrase table with multiple scores That s what you will get from Moses: Swedish English Scores, det?, it s att bli besvikna be disappointed att bli en sj?v to becoming one bara vi just bara just bara naught but bara only phrase translation probability φ(f e) lexical weighting lex(f e) phrase translation probability φ(e f ) lexical weighting lex(e f ) phrase penalty (always exp(1) 2.718) Jörg Tiedemann 28/1 Jörg Tiedemann 29/1
8 Translation = decoding Global search: Ê = argmax EP(E F ) Maria no dio una bofetada a la many translation alternatives (huge phrase table) many ways to segment words into phrases re-ordering makes it even more complex Very Expensive! need search heuristics pruning (early discard weak hypotheses) stack decoding (histograms & thresholds) reordering limits Mary build translation left-to-right select foreign word to be translated select translation in phrase table add translation to partial translation (hypothesis) Jörg Tiedemann 30/1 Jörg Tiedemann 31/1 Maria no dio una bofetada a la Maria no dio una bofetada a la Mary did not Mary did not slap mark first (foreign) word as translated new example: one-to-many translation many-to-one translation Jörg Tiedemann 32/1 Jörg Tiedemann 33/1
9 Maria no dio una bofetada a la Maria no dio una bofetada a la Mary did not slap the Mary did not slap the green many-to-one translation example for re-ordering Jörg Tiedemann 34/1 Jörg Tiedemann 35/1 Lattice of translation options Maria no dio una bofetada a la Mary did not slap the green witch translation finished Jörg Tiedemann 36/1 Jörg Tiedemann 37/1
10 Hypothesis expansion Hypothesis expansion Jörg Tiedemann 38/1 Jörg Tiedemann 39/1 Hypothesis expansion Hypothesis Stacks... and continue adding more hypothesis exponential explosion of search space! here: based on number of foreign words translated expand all hypotheses from one stack during translation place expanded hypotheses into appropriate stacks get n-best list of translations Jörg Tiedemann 40/1 Jörg Tiedemann 41/1
11 Phrase-based SMT Summary PB-SMT More information: Homepage of the Moses toolkit phrase-based SMT = state-of-the-art in data-driven MT (?!) based on standard word alignment models phrase extraction heuristics & simple scoring simplistic re-ordering model huge phrase table = big memory of fragment translations heuristics for efficient decoding Active research area! New developments all the time! Jörg Tiedemann 42/1 Jörg Tiedemann 43/1 What s next? Next lab session: build your own parallel corpus sentence & word alignment Lecture: a quick look at other topics course summary Last lab session: small-scale experiments with PB-SMT (Moses) basic training, evaluation shifting domains Jörg Tiedemann 44/1
The revolution of the empiricists. Machine Translation. Motivation for Data-Driven MT. Machine Translation as Search
The revolution of the empiricists Machine Translation Word alignment & Statistical MT Jörg Tiedemann jorg.tiedemann@lingfil.uu.se Department of Linguistics and Philology Uppsala University Classical approaches
More informationMachine Translation - Decoding
January 15, 2007 Table of Contents 1 Introduction 2 3 4 5 6 Integer Programing Decoder 7 Experimental Results Word alignments Fertility Table Translation Table Heads Non-heads NULL-generated (ct.) Figure:
More informationChallenges in Statistical Machine Translation
p.1 Challenges in Statistical Machine Translation Philipp Koehn koehn@csail.mit.edu Computer Science and Artificial Intelligence Lab Massachusetts Institute of Technology Outline p Statistical Machine
More informationYu Chen Andreas Eisele Martin Kay
LREC 2008: Marrakech, Morocco Department of Computational Linguistics Saarland University May 29, 2008 Outline 1 2 3 4 5 Outline 1 2 3 4 5 SMT architecture To build a phrase-based SMT system: Parallel
More information신경망기반자동번역기술. Konkuk University Computational Intelligence Lab. 김강일
신경망기반자동번역기술 Konkuk University Computational Intelligence Lab. http://ci.konkuk.ac.kr kikim01@kunkuk.ac.kr 김강일 Index Issues in AI and Deep Learning Overview of Machine Translation Advanced Techniques in
More informationStatistical Machine Translation with Long Phrase Table and without Long Parallel Sentences
Statistical Machine Translation with Long Phrase Table and without Long Parallel Sentences Jin ichi Murakami, Masato Tokuhisa, Satoru Ikehara Department of Information and Knowledge Engineering Faculty
More informationMidterm for Name: Good luck! Midterm page 1 of 9
Midterm for 6.864 Name: 40 30 30 30 Good luck! 6.864 Midterm page 1 of 9 Part #1 10% We define a PCFG where the non-terminals are {S, NP, V P, V t, NN, P P, IN}, the terminal symbols are {Mary,ran,home,with,John},
More informationIntroduction to Markov Models
Introduction to Markov Models But first: A few preliminaries Estimating the probability of phrases of words, sentences, etc. CIS 391 - Intro to AI 2 What counts as a word? A tricky question. How to find
More informationCSCI 5832 Natural Language Processing
CSCI 5832 Natural Language Processing Lecture 25 Jim Martin 4/24/07 CSCI 5832 Spring 2007 1 Machine Translation Slides stolen from Kevin Knight (USC/ISI) 4/24/07 CSCI 5832 Spring 2007 2 1 Today 4/24 Machine
More informationA Two-step Technique for MRI Audio Enhancement Using Dictionary Learning and Wavelet Packet Analysis
A Two-step Technique for MRI Audio Enhancement Using Dictionary Learning and Wavelet Packet Analysis Colin Vaz, Vikram Ramanarayanan, and Shrikanth Narayanan USC SAIL Lab INTERSPEECH Articulatory Data
More informationRule Filtering by Pattern for Efficient Hierarchical Translation
for Efficient Hierarchical Translation Gonzalo Iglesias 1 Adrià de Gispert 2 Eduardo R. Banga 1 William Byrne 2 1 Department of Signal Processing and Communications University of Vigo, Spain 2 Department
More informationLog-linear models (part 1I)
Log-linear models (part 1I) Lecture, Feb 2 CS 690N, Spring 2017 Advanced Natural Language Processing http://people.cs.umass.edu/~brenocon/anlp2017/ Brendan O Connor College of Information and Computer
More informationLecture 4: n-grams in NLP. LING 1330/2330: Introduction to Computational Linguistics Na-Rae Han
Lecture 4: n-grams in NLP LING 1330/2330: Introduction to Computational Linguistics Na-Rae Han Objectives Frequent n-grams in English n-grams and statistical NLP n-grams and conditional probability Large
More informationMachine Learning for Language Technology
Machine Learning for Language Technology Generative and Discriminative Models Joakim Nivre Uppsala University Department of Linguistics and Philology joakim.nivre@lingfil.uu.se Machine Learning for Language
More informationIntroduction to Markov Models. Estimating the probability of phrases of words, sentences, etc.
Introduction to Markov Models Estimating the probability of phrases of words, sentences, etc. But first: A few preliminaries on text preprocessing What counts as a word? A tricky question. CIS 421/521
More informationLog-linear models (part 1I)
Log-linear models (part 1I) CS 690N, Spring 2018 Advanced Natural Language Processing http://people.cs.umass.edu/~brenocon/anlp2018/ Brendan O Connor College of Information and Computer Sciences University
More informationThe challenge of simultaneous speech translation
The challenge of simultaneous speech translation Anoop Sarkar School of Computing Science Simon Fraser University Vancouver, British Columbia, Canada PACLIC 30: Seoul. Oct 30, 2016 1 Simultaneous Translation
More informationUser Goal Change Model for Spoken Dialog State Tracking
User Goal Change Model for Spoken Dialog State Tracking Yi Ma Department of Computer Science & Engineering The Ohio State University Columbus, OH 43210, USA may@cse.ohio-state.edu Abstract In this paper,
More informationMATHEMATICAL MODELS Vol. I - Measurements in Mathematical Modeling and Data Processing - William Moran and Barbara La Scala
MEASUREMENTS IN MATEMATICAL MODELING AND DATA PROCESSING William Moran and University of Melbourne, Australia Keywords detection theory, estimation theory, signal processing, hypothesis testing Contents.
More informationDiscriminative Training for Automatic Speech Recognition
Discriminative Training for Automatic Speech Recognition 22 nd April 2013 Advanced Signal Processing Seminar Article Heigold, G.; Ney, H.; Schluter, R.; Wiesler, S. Signal Processing Magazine, IEEE, vol.29,
More informationPart of Speech Tagging & Hidden Markov Models (Part 1) Mitch Marcus CIS 421/521
Part of Speech Tagging & Hidden Markov Models (Part 1) Mitch Marcus CIS 421/521 NLP Task I Determining Part of Speech Tags Given a text, assign each token its correct part of speech (POS) tag, given its
More information/665 Natural Language Processing
601.465/665 Natural Language Processing Prof: Jason Eisner Webpage: http://cs.jhu.edu/~jason/465 syllabus, announcements, slides, homeworks 1 Goals of the field Computers would be a lot more useful if
More informationCS 188: Artificial Intelligence Spring Speech in an Hour
CS 188: Artificial Intelligence Spring 2006 Lecture 19: Speech Recognition 3/23/2006 Dan Klein UC Berkeley Many slides from Dan Jurafsky Speech in an Hour Speech input is an acoustic wave form s p ee ch
More informationRecap from previous lecture. Information Retrieval. Topics for Today. Recall: Basic structure of an Inverted index. Dictionaries & Tolerant Retrieval
Recap from previous lecture nformation Retrieval Dictionaries & Tolerant Retrieval Jörg Tiedemann jorg.tiedemann@lingfil.uu.se Department of Linguistics and Philology Uppsala University nverted indexes
More informationLecture 9b Convolutional Coding/Decoding and Trellis Code modulation
Lecture 9b Convolutional Coding/Decoding and Trellis Code modulation Convolutional Coder Basics Coder State Diagram Encoder Trellis Coder Tree Viterbi Decoding For Simplicity assume Binary Sym.Channel
More informationSample Spaces, Events, Probability
Sample Spaces, Events, Probability CS 3130/ECE 3530: Probability and Statistics for Engineers August 28, 2014 Sets A set is a collection of unique objects. Sets A set is a collection of unique objects.
More informationRecap from previous lectures. Information Retrieval. Recap from previous lectures. Topics for Today. Dictionaries & Tolerant Retrieval.
Recap from previous lectures nformation Retrieval Dictionaries & Tolerant Retrieval Jörg Tiedemann jorg.tiedemann@lingfil.uu.se Department of Linguistics and Philology Uppsala University nverted indexes
More informationTeddy Mantoro.
Teddy Mantoro Email: teddy@ieee.org 1. Title and Abstract 2. AI Method 3. Induction Approach 4. Writing Abstract 5. Writing Introduction What should be in the title: Problem, Method and Result The title
More informationCS 540: Introduction to Artificial Intelligence
CS 540: Introduction to Artificial Intelligence Mid Exam: 7:15-9:15 pm, October 25, 2000 Room 1240 CS & Stats CLOSED BOOK (one sheet of notes and a calculator allowed) Write your answers on these pages
More informationIBM Research Report. Audits and Business Controls Related to Receipt Rules: Benford's Law and Beyond
RC24491 (W0801-103) January 25, 2008 Other IBM Research Report Audits and Business Controls Related to Receipt Rules: Benford's Law and Beyond Vijay Iyengar IBM Research Division Thomas J. Watson Research
More informationPrecoding and Signal Shaping for Digital Transmission
Precoding and Signal Shaping for Digital Transmission Robert F. H. Fischer The Institute of Electrical and Electronics Engineers, Inc., New York WILEY- INTERSCIENCE A JOHN WILEY & SONS, INC., PUBLICATION
More informationSpeech Synthesis using Mel-Cepstral Coefficient Feature
Speech Synthesis using Mel-Cepstral Coefficient Feature By Lu Wang Senior Thesis in Electrical Engineering University of Illinois at Urbana-Champaign Advisor: Professor Mark Hasegawa-Johnson May 2018 Abstract
More informationTeddy Mantoro.
Teddy Mantoro Email: teddy@ieee.org Marshal D Carper Hannah Heath The secret of good writing is rewriting The secret of rewriting is rethinking 1. Title and Abstract 2. AI Method 3. Induction Approach
More informationAI Plays Yun Nie (yunn), Wenqi Hou (wenqihou), Yicheng An (yicheng)
AI Plays 2048 Yun Nie (yunn), Wenqi Hou (wenqihou), Yicheng An (yicheng) Abstract The strategy game 2048 gained great popularity quickly. Although it is easy to play, people cannot win the game easily,
More informationMODIFIED DCT BASED SPEECH ENHANCEMENT IN VEHICULAR ENVIRONMENTS
MODIFIED DCT BASED SPEECH ENHANCEMENT IN VEHICULAR ENVIRONMENTS 1 S.PRASANNA VENKATESH, 2 NITIN NARAYAN, 3 K.SAILESH BHARATHWAAJ, 4 M.P.ACTLIN JEEVA, 5 P.VIJAYALAKSHMI 1,2,3,4,5 SSN College of Engineering,
More informationGrounding into bits: the semantics of virtual worlds
Grounding into bits: the semantics of virtual worlds CHRIS QUIRK /// UW MSR SUMMER INSTITUTE /// 2013 JULY 23 JOINT WORK WITH BILL DOLAN, CHRIS BROCKETT, PALLAVI CHOUDHURY, LUKE ZETTLEMOYER, SVITLANA VOLKOVA,
More informationChapter 4 SPEECH ENHANCEMENT
44 Chapter 4 SPEECH ENHANCEMENT 4.1 INTRODUCTION: Enhancement is defined as improvement in the value or Quality of something. Speech enhancement is defined as the improvement in intelligibility and/or
More informationIntroduction. Description of the Project. Debopam Das
Computational Analysis of Text Sentiment: A Report on Extracting Contextual Information about the Occurrence of Discourse Markers Debopam Das Introduction This report documents a particular task performed
More informationREPORT ITU-R BO Multiple-feed BSS receiving antennas
Rep. ITU-R BO.2102 1 REPORT ITU-R BO.2102 Multiple-feed BSS receiving antennas (2007) 1 Introduction This Report addresses technical and performance issues associated with the design of multiple-feed BSS
More informationDetection of Compound Structures in Very High Spatial Resolution Images
Detection of Compound Structures in Very High Spatial Resolution Images Selim Aksoy Department of Computer Engineering Bilkent University Bilkent, 06800, Ankara, Turkey saksoy@cs.bilkent.edu.tr Joint work
More informationELT Receiver Architectures and Signal Processing Fall Mandatory homework exercises
ELT-44006 Receiver Architectures and Signal Processing Fall 2014 1 Mandatory homework exercises - Individual solutions to be returned to Markku Renfors by email or in paper format. - Solutions are expected
More informationThe fundamentals of detection theory
Advanced Signal Processing: The fundamentals of detection theory Side 1 of 18 Index of contents: Advanced Signal Processing: The fundamentals of detection theory... 3 1 Problem Statements... 3 2 Detection
More informationVQ Source Models: Perceptual & Phase Issues
VQ Source Models: Perceptual & Phase Issues Dan Ellis & Ron Weiss Laboratory for Recognition and Organization of Speech and Audio Dept. Electrical Eng., Columbia Univ., NY USA {dpwe,ronw}@ee.columbia.edu
More informationLearning Structured Predictors
Learning Structured Predictors Xavier Carreras Xerox Research Centre Europe Supervised (Structured) Prediction Learning to predict: given training data { (x (1), y (1) ), (x (2), y (2) ),..., (x (m), y
More informationCandyCrush.ai: An AI Agent for Candy Crush
CandyCrush.ai: An AI Agent for Candy Crush Jiwoo Lee, Niranjan Balachandar, Karan Singhal December 16, 2016 1 Introduction Candy Crush, a mobile puzzle game, has become very popular in the past few years.
More information2048: An Autonomous Solver
2048: An Autonomous Solver Final Project in Introduction to Artificial Intelligence ABSTRACT. Our goal in this project was to create an automatic solver for the wellknown game 2048 and to analyze how different
More informationTechniques for Sentiment Analysis survey
I J C T A, 9(41), 2016, pp. 355-360 International Science Press ISSN: 0974-5572 Techniques for Sentiment Analysis survey Anu Sharma* and Savleen Kaur** ABSTRACT A Sentiment analysis is a technique to analyze
More informationWhy Should We Care? Everyone uses plotting But most people ignore or are unaware of simple principles Default plotting tools are not always the best
Elementary Plots Why Should We Care? Everyone uses plotting But most people ignore or are unaware of simple principles Default plotting tools are not always the best More importantly, it is easy to lie
More informationAI Approaches to Ultimate Tic-Tac-Toe
AI Approaches to Ultimate Tic-Tac-Toe Eytan Lifshitz CS Department Hebrew University of Jerusalem, Israel David Tsurel CS Department Hebrew University of Jerusalem, Israel I. INTRODUCTION This report is
More informationA Bit of network information theory
Š#/,% 0/,94%#(.)15% A Bit of network information theory Suhas Diggavi 1 Email: suhas.diggavi@epfl.ch URL: http://licos.epfl.ch Parts of talk are joint work with S. Avestimehr 2, S. Mohajer 1, C. Tian 3,
More informationCheap, Fast and Good Enough: Speech Transcription with Mechanical Turk. Scott Novotney and Chris Callison-Burch 04/02/10
Cheap, Fast and Good Enough: Speech Transcription with Mechanical Turk Scott Novotney and Chris Callison-Burch 04/02/10 Motivation Speech recognition models hunger for data ASR requires thousands of hours
More informationPatent Mining: Use of Data/Text Mining for Supporting Patent Retrieval and Analysis
Patent Mining: Use of Data/Text Mining for Supporting Patent Retrieval and Analysis by Chih-Ping Wei ( 魏志平 ), PhD Institute of Service Science and Institute of Technology Management National Tsing Hua
More informationWireless Network Coding with Local Network Views: Coded Layer Scheduling
Wireless Network Coding with Local Network Views: Coded Layer Scheduling Alireza Vahid, Vaneet Aggarwal, A. Salman Avestimehr, and Ashutosh Sabharwal arxiv:06.574v3 [cs.it] 4 Apr 07 Abstract One of the
More informationTHE goal of Speaker Diarization is to segment audio
SUBMITTED TO IEEE TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING 1 The ICSI RT-09 Speaker Diarization System Gerald Friedland* Member IEEE, Adam Janin, David Imseng Student Member IEEE, Xavier
More informationUsing RASTA in task independent TANDEM feature extraction
R E S E A R C H R E P O R T I D I A P Using RASTA in task independent TANDEM feature extraction Guillermo Aradilla a John Dines a Sunil Sivadas a b IDIAP RR 04-22 April 2004 D a l l e M o l l e I n s t
More informationLecture 2: SIGNALS. 1 st semester By: Elham Sunbu
Lecture 2: SIGNALS 1 st semester 1439-2017 1 By: Elham Sunbu OUTLINE Signals and the classification of signals Sine wave Time and frequency domains Composite signals Signal bandwidth Digital signal Signal
More informationProgress in the BBN Keyword Search System for the DARPA RATS Program
INTERSPEECH 2014 Progress in the BBN Keyword Search System for the DARPA RATS Program Tim Ng 1, Roger Hsiao 1, Le Zhang 1, Damianos Karakos 1, Sri Harish Mallidi 2, Martin Karafiát 3,KarelVeselý 3, Igor
More informationIntroduction to HTK Toolkit
Introduction to HTK Toolkit Berlin Chen 2004 Reference: - Steve Young et al. The HTK Book. Version 3.2, 2002. Outline An Overview of HTK HTK Processing Stages Data Preparation Tools Training Tools Testing
More informationIntroduction to probability
Introduction to probability Suppose an experiment has a finite set X = {x 1,x 2,...,x n } of n possible outcomes. Each time the experiment is performed exactly one on the n outcomes happens. Assign each
More informationNLP, Games, and Robotic Cars
NLP, Games, and Robotic Cars [These slides were created by Dan Klein and Pieter Abbeel for CS188 Intro to AI at UC Berkeley. All CS188 materials are available at http://ai.berkeley.edu.] So Far: Foundational
More informationLecture 5: Pitch and Chord (1) Chord Recognition. Li Su
Lecture 5: Pitch and Chord (1) Chord Recognition Li Su Recap: short-time Fourier transform Given a discrete-time signal x(t) sampled at a rate f s. Let window size N samples, hop size H samples, then the
More informationApplication of QAP in Modulation Diversity (MoDiv) Design
Application of QAP in Modulation Diversity (MoDiv) Design Hans D Mittelmann School of Mathematical and Statistical Sciences Arizona State University INFORMS Annual Meeting Philadelphia, PA 4 November 2015
More informationA Bandit Approach for Tree Search
A An Example in Computer-Go Department of Statistics, University of Michigan March 27th, 2008 A 1 Bandit Problem K-Armed Bandit UCB Algorithms for K-Armed Bandit Problem 2 Classical Tree Search UCT Algorithm
More informationA Correlation-Maximization Denoising Filter Used as An Enhancement Frontend for Noise Robust Bird Call Classification
A Correlation-Maximization Denoising Filter Used as An Enhancement Frontend for Noise Robust Bird Call Classification Wei Chu and Abeer Alwan Speech Processing and Auditory Perception Laboratory Department
More informationChapter 1. Probability
Chapter 1. Probability 1.1 Basic Concepts Scientific method a. For a given problem, we define measures that explains the problem well. b. Data is collected with observation and the measures are calculated.
More informationEE 435/535: Error Correcting Codes Project 1, Fall 2009: Extended Hamming Code. 1 Introduction. 2 Extended Hamming Code: Encoding. 1.
EE 435/535: Error Correcting Codes Project 1, Fall 2009: Extended Hamming Code Project #1 is due on Tuesday, October 6, 2009, in class. You may turn the project report in early. Late projects are accepted
More informationTwo Bracketing Schemes for the Penn Treebank
Anssi Yli-Jyrä Two Bracketing Schemes for the Penn Treebank Abstract The trees in the Penn Treebank have a standard representation that involves complete balanced bracketing. In this article, an alternative
More informationIntroduction to Source Coding
Comm. 52: Communication Theory Lecture 7 Introduction to Source Coding - Requirements of source codes - Huffman Code Length Fixed Length Variable Length Source Code Properties Uniquely Decodable allow
More informationCS 343: Artificial Intelligence
CS 343: Artificial Intelligence NLP, Games, and Autonomous Vehicles Prof. Scott Niekum The University of Texas at Austin [These slides based on those of Dan Klein and Pieter Abbeel for CS188 Intro to AI
More informationIMPROVEMENTS TO THE IBM SPEECH ACTIVITY DETECTION SYSTEM FOR THE DARPA RATS PROGRAM
IMPROVEMENTS TO THE IBM SPEECH ACTIVITY DETECTION SYSTEM FOR THE DARPA RATS PROGRAM Samuel Thomas 1, George Saon 1, Maarten Van Segbroeck 2 and Shrikanth S. Narayanan 2 1 IBM T.J. Watson Research Center,
More informationProbability and Statistics. Copyright Cengage Learning. All rights reserved.
Probability and Statistics Copyright Cengage Learning. All rights reserved. 14.2 Probability Copyright Cengage Learning. All rights reserved. Objectives What Is Probability? Calculating Probability by
More informationA Static Power Model for Architects
A Static Power Model for Architects J. Adam Butts and Guri Sohi University of Wisconsin-Madison {butts,sohi}@cs.wisc.edu 33rd International Symposium on Microarchitecture Monterey, California December,
More informationReview: Theorem of irrelevance. Y j φ j (t) where Y j = X j + Z j for 1 j k and Y j = Z j for
Review: Theorem of irrelevance Given the signal set { a 1,..., a M }, we transmit X(t) = j k =1 a m,jφ j (t) and receive Y (t) = j=1 Y j φ j (t) where Y j = X j + Z j for 1 j k and Y j = Z j for j>k. Assume
More informationDSP First Lab 08: Frequency Response: Bandpass and Nulling Filters
DSP First Lab 08: Frequency Response: Bandpass and Nulling Filters Pre-Lab and Warm-Up: You should read at least the Pre-Lab and Warm-up sections of this lab assignment and go over all exercises in the
More informationDatabase Normalization as a By-product of MML Inference. Minimum Message Length Inference
Database Normalization as a By-product of Minimum Message Length Inference David Dowe Nayyar A. Zaidi Clayton School of IT, Monash University, Melbourne VIC 3800, Australia December 8, 2010 Our Research
More informationCHANNEL MEASUREMENT. Channel measurement doesn t help for single bit transmission in flat Rayleigh fading.
CHANNEL MEASUREMENT Channel measurement doesn t help for single bit transmission in flat Rayleigh fading. It helps (as we soon see) in detection with multi-tap fading, multiple frequencies, multiple antennas,
More informationTranscribing Continuous Speech Using Mismatched Crowdsourcing
Transcribing Continuous Speech Using Mismatched Crowdsourcing Preethi Jyothi 1, Mark Hasegawa-Johnson 1,2 1 Beckman Institute, University of Illinois at Urbana-Champaign, US 2 Department of ECE, University
More informationMAS160: Signals, Systems & Information for Media Technology. Problem Set 4. DUE: October 20, 2003
MAS160: Signals, Systems & Information for Media Technology Problem Set 4 DUE: October 20, 2003 Instructors: V. Michael Bove, Jr. and Rosalind Picard T.A. Jim McBride Problem 1: Simple Psychoacoustic Masking
More informationThe Case for Optimum Detection Algorithms in MIMO Wireless Systems. Helmut Bölcskei
The Case for Optimum Detection Algorithms in MIMO Wireless Systems Helmut Bölcskei joint work with A. Burg, C. Studer, and M. Borgmann ETH Zurich Data rates in wireless double every 18 months throughput
More informationElectric Guitar Pickups Recognition
Electric Guitar Pickups Recognition Warren Jonhow Lee warrenjo@stanford.edu Yi-Chun Chen yichunc@stanford.edu Abstract Electric guitar pickups convert vibration of strings to eletric signals and thus direcly
More informationWhy Should We Care? More importantly, it is easy to lie or deceive people with bad plots
Elementary Plots Why Should We Care? Everyone uses plotting But most people ignore or are unaware of simple principles Default plotting tools (or default settings) are not always the best More importantly,
More informationWavelets and wavelet convolution and brain music. Dr. Frederike Petzschner Translational Neuromodeling Unit
Wavelets and wavelet convolution and brain music Dr. Frederike Petzschner Translational Neuromodeling Unit 06.03.2015 Recap Why are we doing this? We know that EEG data contain oscillations. Or goal is
More informationSPEECH TO SINGING SYNTHESIS SYSTEM. Mingqing Yun, Yoon mo Yang, Yufei Zhang. Department of Electrical and Computer Engineering University of Rochester
SPEECH TO SINGING SYNTHESIS SYSTEM Mingqing Yun, Yoon mo Yang, Yufei Zhang Department of Electrical and Computer Engineering University of Rochester ABSTRACT This paper describes a speech-to-singing synthesis
More informationSimple Measures of Visual Encoding. vs. Information Theory
Simple Measures of Visual Encoding vs. Information Theory Simple Measures of Visual Encoding STIMULUS RESPONSE What does a [visual] neuron do? Tuning Curves Receptive Fields Average Firing Rate (Hz) Stimulus
More informationSELECTING RELEVANT DATA
EXPLORATORY ANALYSIS The data that will be used comes from the reviews_beauty.json.gz file which contains information about beauty products that were bought and reviewed on Amazon.com. Each data point
More informationPlayer Speed vs. Wild Pokémon Encounter Frequency in Pokémon SoulSilver Joshua and AP Statistics, pd. 3B
Player Speed vs. Wild Pokémon Encounter Frequency in Pokémon SoulSilver Joshua and AP Statistics, pd. 3B In the newest iterations of Nintendo s famous Pokémon franchise, Pokémon HeartGold and SoulSilver
More informationLECTURE VI: LOSSLESS COMPRESSION ALGORITHMS DR. OUIEM BCHIR
1 LECTURE VI: LOSSLESS COMPRESSION ALGORITHMS DR. OUIEM BCHIR 2 STORAGE SPACE Uncompressed graphics, audio, and video data require substantial storage capacity. Storing uncompressed video is not possible
More informationTh P6 01 Retrieval of the P- and S-velocity Structure of the Groningen Gas Reservoir Using Noise Interferometry
Th P6 1 Retrieval of the P- and S-velocity Structure of the Groningen Gas Reservoir Using Noise Interferometry W. Zhou* (Utrecht University), H. Paulssen (Utrecht University) Summary The Groningen gas
More informationHANDS-ON TRANSFORMATIONS: RIGID MOTIONS AND CONGRUENCE (Poll Code 39934)
HANDS-ON TRANSFORMATIONS: RIGID MOTIONS AND CONGRUENCE (Poll Code 39934) Presented by Shelley Kriegler President, Center for Mathematics and Teaching shelley@mathandteaching.org Fall 2014 8.F.1 8.G.1a
More informationDyck paths, standard Young tableaux, and pattern avoiding permutations
PU. M. A. Vol. 21 (2010), No.2, pp. 265 284 Dyck paths, standard Young tableaux, and pattern avoiding permutations Hilmar Haukur Gudmundsson The Mathematics Institute Reykjavik University Iceland e-mail:
More informationCIS 2033 Lecture 6, Spring 2017
CIS 2033 Lecture 6, Spring 2017 Instructor: David Dobor February 2, 2017 In this lecture, we introduce the basic principle of counting, use it to count subsets, permutations, combinations, and partitions,
More informationCMath 55 PROFESSOR KENNETH A. RIBET. Final Examination May 11, :30AM 2:30PM, 100 Lewis Hall
CMath 55 PROFESSOR KENNETH A. RIBET Final Examination May 11, 015 11:30AM :30PM, 100 Lewis Hall Please put away all books, calculators, cell phones and other devices. You may consult a single two-sided
More informationDigital Speech Processing and Coding
ENEE408G Spring 2006 Lecture-2 Digital Speech Processing and Coding Spring 06 Instructor: Shihab Shamma Electrical & Computer Engineering University of Maryland, College Park http://www.ece.umd.edu/class/enee408g/
More informationYour Name and ID. (a) ( 3 points) Breadth First Search is complete even if zero step-costs are allowed.
1 UC Davis: Winter 2003 ECS 170 Introduction to Artificial Intelligence Final Examination, Open Text Book and Open Class Notes. Answer All questions on the question paper in the spaces provided Show all
More informationSpeech Coding in the Frequency Domain
Speech Coding in the Frequency Domain Speech Processing Advanced Topics Tom Bäckström Aalto University October 215 Introduction The speech production model can be used to efficiently encode speech signals.
More informationCombining Voice Activity Detection Algorithms by Decision Fusion
Combining Voice Activity Detection Algorithms by Decision Fusion Evgeny Karpov, Zaur Nasibov, Tomi Kinnunen, Pasi Fränti Speech and Image Processing Unit, University of Eastern Finland, Joensuu, Finland
More informationAdvanced Digital Design
Advanced Digital Design Introduction & Motivation by A. Steininger and M. Delvai Vienna University of Technology Outline Challenges in Digital Design The Role of Time in the Design The Fundamental Design
More informationMining for Statistical Models of Availability in Large-Scale Distributed Systems: An Empirical Study of
Mining for Statistical Models of Availability in Large-Scale Distributed Systems: An Empirical Study of SETI@home Bahman Javadi 1, Derrick Kondo 1, Jean-Marc Vincent 1,2, David P. Anderson 3 1 Laboratoire
More informationAM Antenna Computer Modeling Course
AM Antenna Computer Modeling Course Course Description The FCC now permits moment method computer modeling of many AM directional arrays as an alternative to traditional cut-and-try adjustments and field
More informationNotes 15: Concatenated Codes, Turbo Codes and Iterative Processing
16.548 Notes 15: Concatenated Codes, Turbo Codes and Iterative Processing Outline! Introduction " Pushing the Bounds on Channel Capacity " Theory of Iterative Decoding " Recursive Convolutional Coding
More information