Timbre Similarity. Perception and Computation. Prof. Michael Casey. Dartmouth College. Thursday 7th February, 2008
|
|
- Kelly Wilkins
- 6 years ago
- Views:
Transcription
1 Timbre Similarity Perception and Computation Prof. Michael Casey Dartmouth College Thursday 7th February, 2008 Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
2 Audio Similarity Computation Metric Spaces Euclidean and Cosine Metrics The S-Matrix Timbre Spaces Sound Objects and Textures Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
3 Metric Spaces A metric space is a vector space with a distance function, called a metric. a, b R d L p norm: δ p (a, b) = { d i=1 (a i b i ) p } 1 p L 1 is the City Block metric L 2 is the Euclidean distance Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
4 Euclidean and Cosine Distance Euclidean distance converts rectangular coordinates to magnitude (length). Cosine distance converts rectangular coordinates to the cosine of the angle between vectors. Dot product: ab T = d i=1 a i b i Cosine distance: cos(θ) = abt a b If a = 1 and b = 1 then cos(θ) = ab T If a = 1 and b = 1 then δ 2 (a, b) = 2 2ab T Else: δ 2 (a, b) = a + b 2 a b abt Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
5 Audio Feature Vectors We use a feature extractor to compute a set of feature vectors. We obtain a new vector of dimensionality d every N samples. The collection of vectors forms an observation matrix X R t d x 11 x 12 x x 1d x 21 x 22 x x 2d X = x t1 x t2 x t3... x td Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
6 Feature Vector Norming Often we want to make the features invariant to scaling. To do this we make each vector unit norm: ˆx = x x = x d i=1 x2 i By doing this we make Euclidean distance proportional to cosine distance: δ 2 (ˆx, ŷ) = 2 2ˆxŷ T Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
7 Self-Similarity Matrix: S-Matrix Jonathan Foote, Visualizing music and audio using self-similarity, Proceedings of the seventh ACM international conference on Multimedia, Orlando, S-Matrix is based on the cosine distance between normed matrices: S = ˆXˆX T We implement the S-Matrix in Octave using matrix multiplication: octave> X = loadadb( myfeatures.mfcc20 ); octave> X = nmmtx(x); octave> S = X*X ; octave> imagesc(s); Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
8 S-Matrix: Chopin Mazurka Opus 6 No MFCC Coefficients; 100ms hop; normed cosine distance. Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
9 Audio Feature Extraction fftextract can be downloaded from It is easy to extract features for.wav,.aiff,.ogg,.snd files using this program. If you have compressed audio :.mp3,.mp4,.aac,.wma,.flac you will first need to decode them to PCM format. Also download the following files and put them in your Octave/Matlab folder: nmmtx.m Unit norm each vector in an observation matrix dist.m Euclidean distance between observation matrices readadb.m Load an observation matrix Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
10 fftextract Feature Example: 12 band-per-octave Constant-Q spectrum: fftextract -q 12 file.wav file.cqt12 Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
11 fftextract Feature Example: 24-band Pitch-Class Profile (PCP) fftextract -c 24 file.wav file.chr24 Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
12 fftextract Feature Example: 20 Mel-frequency cepstral coefficients (MFCC) fftextract -m 20 file.wav file.mfcc20 Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
13 Sound Object Similarity Sound objects are short individual events. Make an observation matrix out of sound object features. Each sound object s features forms one row of the matrix. The S-Matrix is a distance matrix of each sound to each other sound. Can we derive a map of the similarity space between sounds? Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
14 Kruskal s Multidimensional Scaling (MDS) Algorithm MDS takes a matrix of distances as input. The output is a map of points in d-dimensional space d is chosen to be as low as possible with minimal Kruskal stress. Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
15 Distances between 10 major UK Cities Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
16 MDS Example: recovering the map of the UK Measure the distances between cities in your favourite country Make a symmetric distance matrix of these distances (like in your AAA books) Run the MDS algorithm on the distance matrix The result is a recovered map of the positions MDS Solution for 10 Cities in the UK Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
17 Sound Object Similarity Which sound objects are perceived as similar? Is sound perception continuous, categorical or both? How can we compute the similarity of two sounds? How can we compute the category of a sound? What identifies a sound as belonging to a category? Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
18 Timbre Space of Musical Instrument Relationships Make a collection of features for a set of musical instruments. Fix: volume / duration / pitch Measure the distances using an S-Matrix. Run the MDS algorithm on the distance matrix The result is a recovered map of the positions in a timbre space. Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
19 Texture Similarity Music recordings and everyday sounds consist of sound textures; Textures are simultaneous sounding events. It is common to think of textures as noisy but counterpoint is a texture. The same questions apply to sound textures as for sound objects: Which textures are perceived as similar? Is texture perception continuous, categorical or both? How can we compute the similarity of sound textures? How can we compute the category of a sound texture? What identifies a sound texture belonging to a category? Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
20 Musical Instrument Timbre Perception John Grey (1975) Multidimensional Scaling (MDS) of Musical Instruments David Wessel (1979) Perceptual Control Spaces Jean-Claude Risset (1979) MDS of Re-Synthesized Instrument Tones Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
21 Everyday Audio Perception Bill Gaver (1983) Ecological Audio Perception Warren and Vebrugge (1989) Perception of Breaking-Bouncing Events Clarkson (1995) Classification of Ambulatory Audio Prof. Michael Casey (Dartmouth College) Timbre Similarity Thursday 7th February, / 21
University of Colorado at Boulder ECEN 4/5532. Lab 2 Lab report due on February 16, 2015
University of Colorado at Boulder ECEN 4/5532 Lab 2 Lab report due on February 16, 2015 This is a MATLAB only lab, and therefore each student needs to turn in her/his own lab report and own programs. 1
More informationIntroduction Basic Audio Feature Extraction
Introduction Basic Audio Feature Extraction Vincent Koops (with slides by Meinhard Müller) Sound and Music Technology, December 6th, 2016 1 28 November 2017 Today g Main modules A. Sound and music for
More informationTopic 6. Timbre Representations
Topic 6 Timbre Representations We often say that singer s voice is magnetic the violin sounds bright this French horn sounds solid that drum sounds dull What aspect(s) of sound are these words describing?
More informationMultimedia Retrieval Distance. Egon L. van den Broek
Multimedia Retrieval 2018-1019 Distance Egon L. van den Broek 1 The project: Two perspectives Man Machine or? Objective Subjective 2 The default Default: distance = Euclidean distance This is how it is
More informationShort-Time Fourier Transform and Chroma Features
Friedrich-Alexander-Universität Erlangen-Nürnberg Lab Course Short-Time Fourier Transform and Chroma Features International Audio Laboratories Erlangen Prof. Dr. Meinard Müller Friedrich-Alexander Universität
More informationMDS codec evaluation based on perceptual sound attributes
MDS codec evaluation based on perceptual sound attributes Marcelo Herrera Martínez * Edwar Jacinto Gómez ** Edilberto Carlos Vivas G. *** submitted date: March 03 received date: April 03 accepted date:
More informationBASIC COMPRESSION TECHNIQUES
BASIC COMPRESSION TECHNIQUES N. C. State University CSC557 Multimedia Computing and Networking Fall 2001 Lectures # 05 Questions / Problems / Announcements? 2 Matlab demo of DFT Low-pass windowed-sinc
More informationDominant Feature Vectors Based Audio Similarity Measure
Dominant Feature Vectors Based Audio Similarity Measure Jing Gu 1, Lie Lu 2, Rui Cai 3, Hong-Jiang Zhang 2, and Jian Yang 1 1 Dept. of Electronic Engineering, Tsinghua Univ., Beijing, 100084, China 2 Microsoft
More informationDistances and similarities Based in part on slides from textbook, slides of Susan Holmes. October 3, Statistics 202: Data Mining
Distances and similarities Based in part on slides from textbook, slides of Susan Holmes October 3, 2012 1 / 1 Similarities Start with X which we assume is centered and standardized. The PCA loadings were
More informationOn Spectral Basis Selection for Single Channel Polyphonic Music Separation
On Spectral Basis Selection for Single Channel Polyphonic Music Separation Minje Kim and Seungjin Choi Department of Computer Science Pohang University of Science and Technology San 31 Hyoja-dong, Nam-gu
More informationShort-Time Fourier Transform and Chroma Features
Friedrich-Alexander-Universität Erlangen-Nürnberg Lab Course Short-Time Fourier Transform and Chroma Features International Audio Laboratories Erlangen Prof. Dr. Meinard Müller Friedrich-Alexander Universität
More informationAnalysis of polyphonic audio using source-filter model and non-negative matrix factorization
Analysis of polyphonic audio using source-filter model and non-negative matrix factorization Tuomas Virtanen and Anssi Klapuri Tampere University of Technology, Institute of Signal Processing Korkeakoulunkatu
More informationIMISOUND: An Unsupervised System for Sound Query by Vocal Imitation
IMISOUND: An Unsupervised System for Sound Query by Vocal Imitation Yichi Zhang and Zhiyao Duan Audio Information Research (AIR) Lab Department of Electrical and Computer Engineering University of Rochester
More informationMusical Genre Classication
Musical Genre Classication Jan Müllers RWTH Aachen, 2015 Jan Müllers Finding Disjoint Paths 1 / 15 Musical Genres The Problem Musical Genres History Automatic Speech Regocnition categorical labels created
More informationTime-Frequency Analysis
Time-Frequency Analysis Basics of Fourier Series Philippe B. aval KSU Fall 015 Philippe B. aval (KSU) Fourier Series Fall 015 1 / 0 Introduction We first review how to derive the Fourier series of a function.
More informationFINM 331: MULTIVARIATE DATA ANALYSIS FALL 2017 PROBLEM SET 3
FINM 331: MULTIVARIATE DATA ANALYSIS FALL 2017 PROBLEM SET 3 The required files for all problems can be found in: http://www.stat.uchicago.edu/~lekheng/courses/331/hw3/ The file name indicates which problem
More informationChapter 17: Fourier Series
Section A Introduction to Fourier Series By the end of this section you will be able to recognise periodic functions sketch periodic functions determine the period of the given function Why are Fourier
More informationLecture 7: Feature Extraction
Lecture 7: Feature Extraction Kai Yu SpeechLab Department of Computer Science & Engineering Shanghai Jiao Tong University Autumn 2014 Kai Yu Lecture 7: Feature Extraction SJTU Speech Lab 1 / 28 Table of
More informationAudio Features. Fourier Transform. Short Time Fourier Transform. Short Time Fourier Transform. Short Time Fourier Transform
Advanced Course Computer Science Music Processing Summer Term 2009 Meinard Müller Saarland University and MPI Informatik meinard@mpi-inf.mpg.de Audio Features Fourier Transform Tells which notes (frequencies)
More informationStatistics 202: Data Mining. c Jonathan Taylor. Week 2 Based in part on slides from textbook, slides of Susan Holmes. October 3, / 1
Week 2 Based in part on slides from textbook, slides of Susan Holmes October 3, 2012 1 / 1 Part I Other datatypes, preprocessing 2 / 1 Other datatypes Document data You might start with a collection of
More informationPart I. Other datatypes, preprocessing. Other datatypes. Other datatypes. Week 2 Based in part on slides from textbook, slides of Susan Holmes
Week 2 Based in part on slides from textbook, slides of Susan Holmes Part I Other datatypes, preprocessing October 3, 2012 1 / 1 2 / 1 Other datatypes Other datatypes Document data You might start with
More informationc Springer, Reprinted with permission.
Zhijian Yuan and Erkki Oja. A FastICA Algorithm for Non-negative Independent Component Analysis. In Puntonet, Carlos G.; Prieto, Alberto (Eds.), Proceedings of the Fifth International Symposium on Independent
More information1. Vectors.
1. Vectors 1.1 Vectors and Matrices Linear algebra is concerned with two basic kinds of quantities: vectors and matrices. 1.1 Vectors and Matrices Scalars and Vectors - Scalar: a numerical value denoted
More informationCorrespondence Analysis, Cross-Autocorrelation and Clustering in Polyphonic Music
Cocco, C., Bavaud, F. : Correspondence Analysis, Cross-Autocorrelation and Clustering in Polyphonic Music. In: Lausen, B. et al. (Eds.) Data Science, Learning by Latent Structures, and Knowledge Discovery,
More informationLecture 7. Econ August 18
Lecture 7 Econ 2001 2015 August 18 Lecture 7 Outline First, the theorem of the maximum, an amazing result about continuity in optimization problems. Then, we start linear algebra, mostly looking at familiar
More informationCRYSTALLIZATION SONIFICATION OF HIGH-DIMENSIONAL DATASETS
Proceedings of the 22 International Conference on Auditory Display, Kyoto, Japan, July 2 5, 22 CRYSTALLIZATION SONIFICATION OF HIGH-DIMENSIONAL DATASETS T. Hermann Faculty of Technology Bielefeld University,
More informationSTAT 730 Chapter 14: Multidimensional scaling
STAT 730 Chapter 14: Multidimensional scaling Timothy Hanson Department of Statistics, University of South Carolina Stat 730: Multivariate Data Analysis 1 / 16 Basic idea We have n objects and a matrix
More informationFrom Fourier Series to Analysis of Non-stationary Signals - II
From Fourier Series to Analysis of Non-stationary Signals - II prof. Miroslav Vlcek October 10, 2017 Contents Signals 1 Signals 2 3 4 Contents Signals 1 Signals 2 3 4 Contents Signals 1 Signals 2 3 4 Contents
More informationLecture Notes 5: Multiresolution Analysis
Optimization-based data analysis Fall 2017 Lecture Notes 5: Multiresolution Analysis 1 Frames A frame is a generalization of an orthonormal basis. The inner products between the vectors in a frame and
More informationThe geometry of least squares
The geometry of least squares We can think of a vector as a point in space, where the elements of the vector are the coordinates of the point. Consider for example, the following vector s: t = ( 4, 0),
More informationAutomatic Speech Recognition (CS753)
Automatic Speech Recognition (CS753) Lecture 12: Acoustic Feature Extraction for ASR Instructor: Preethi Jyothi Feb 13, 2017 Speech Signal Analysis Generate discrete samples A frame Need to focus on short
More informationSpectral and Textural Feature-Based System for Automatic Detection of Fricatives and Affricates
Spectral and Textural Feature-Based System for Automatic Detection of Fricatives and Affricates Dima Ruinskiy Niv Dadush Yizhar Lavner Department of Computer Science, Tel-Hai College, Israel Outline Phoneme
More informationMetric-based classifiers. Nuno Vasconcelos UCSD
Metric-based classifiers Nuno Vasconcelos UCSD Statistical learning goal: given a function f. y f and a collection of eample data-points, learn what the function f. is. this is called training. two major
More informationSingle Channel Music Sound Separation Based on Spectrogram Decomposition and Note Classification
Single Channel Music Sound Separation Based on Spectrogram Decomposition and Note Classification Hafiz Mustafa and Wenwu Wang Centre for Vision, Speech and Signal Processing (CVSSP) University of Surrey,
More informationData Mining: Data. Lecture Notes for Chapter 2. Introduction to Data Mining
Data Mining: Data Lecture Notes for Chapter 2 Introduction to Data Mining by Tan, Steinbach, Kumar Similarity and Dissimilarity Similarity Numerical measure of how alike two data objects are. Is higher
More informationSpeech Signal Representations
Speech Signal Representations Berlin Chen 2003 References: 1. X. Huang et. al., Spoken Language Processing, Chapters 5, 6 2. J. R. Deller et. al., Discrete-Time Processing of Speech Signals, Chapters 4-6
More informationAudio Features. Fourier Transform. Fourier Transform. Fourier Transform. Short Time Fourier Transform. Fourier Transform.
Advanced Course Computer Science Music Processing Summer Term 2010 Fourier Transform Meinard Müller Saarland University and MPI Informatik meinard@mpi-inf.mpg.de Audio Features Fourier Transform Fourier
More informationLEHNINGER PRINCIPLES OF BIOCHEMISTRY 6TH EDITION TEST BANK PDF
LEHNINGER PRINCIPLES OF BIOCHEMISTRY 6TH EDITION TEST BANK PDF ==> Download: LEHNINGER PRINCIPLES OF BIOCHEMISTRY 6TH EDITION TEST BANK PDF LEHNINGER PRINCIPLES OF BIOCHEMISTRY 6TH EDITION TEST BANK PDF
More informationRecommendation Systems
Recommendation Systems Popularity Recommendation Systems Predicting user responses to options Offering news articles based on users interests Offering suggestions on what the user might like to buy/consume
More informationTIME-DEPENDENT PARAMETRIC AND HARMONIC TEMPLATES IN NON-NEGATIVE MATRIX FACTORIZATION
TIME-DEPENDENT PARAMETRIC AND HARMONIC TEMPLATES IN NON-NEGATIVE MATRIX FACTORIZATION 13 th International Conference on Digital Audio Effects Romain Hennequin, Roland Badeau and Bertrand David Telecom
More informationMachine Recognition of Sounds in Mixtures
Machine Recognition of Sounds in Mixtures Outline 1 2 3 4 Computational Auditory Scene Analysis Speech Recognition as Source Formation Sound Fragment Decoding Results & Conclusions Dan Ellis
More informationDesigning Information Devices and Systems I Fall 2018 Homework 3
Last Updated: 28-9-5 :8 EECS 6A Designing Information Devices and Systems I Fall 28 Homework 3 This homework is due September 4, 28, at 23:59. Self-grades are due September 8, 28, at 23:59. Submission
More informationNoise Robust Isolated Words Recognition Problem Solving Based on Simultaneous Perturbation Stochastic Approximation Algorithm
EngOpt 2008 - International Conference on Engineering Optimization Rio de Janeiro, Brazil, 0-05 June 2008. Noise Robust Isolated Words Recognition Problem Solving Based on Simultaneous Perturbation Stochastic
More informationFeature extraction 2
Centre for Vision Speech & Signal Processing University of Surrey, Guildford GU2 7XH. Feature extraction 2 Dr Philip Jackson Linear prediction Perceptual linear prediction Comparison of feature methods
More informationIn-class exercises. Day 1
Physics 4488/6562: Statistical Mechanics http://www.physics.cornell.edu/sethna/teaching/562/ Material for Week 13 Exercises due Mon Apr 30 Last correction at April 24, 2018, 12:59 pm c 2018, James Sethna,
More informationA NEW DISSIMILARITY METRIC FOR THE CLUSTERING OF PARTIALS USING THE COMMON VARIATION CUE
A NEW DISSIMILARITY METRIC FOR THE CLUSTERING OF PARTIALS USING THE COMMON VARIATION CUE Mathieu Lagrange SCRIME LaBRI, Université Bordeaux 1 351, cours de la Libération, F-33405 Talence cedex, France
More informationSinger Identification using MFCC and LPC and its comparison for ANN and Naïve Bayes Classifiers
Singer Identification using MFCC and LPC and its comparison for ANN and Naïve Bayes Classifiers Kumari Rambha Ranjan, Kartik Mahto, Dipti Kumari,S.S.Solanki Dept. of Electronics and Communication Birla
More informationRobust Speaker Identification
Robust Speaker Identification by Smarajit Bose Interdisciplinary Statistical Research Unit Indian Statistical Institute, Kolkata Joint work with Amita Pal and Ayanendranath Basu Overview } } } } } } }
More informationencoding without prediction) (Server) Quantization: Initial Data 0, 1, 2, Quantized Data 0, 1, 2, 3, 4, 8, 16, 32, 64, 128, 256
General Models for Compression / Decompression -they apply to symbols data, text, and to image but not video 1. Simplest model (Lossless ( encoding without prediction) (server) Signal Encode Transmit (client)
More informationCSCI 239 Discrete Structures of Computer Science Lab 6 Vectors and Matrices
CSCI 239 Discrete Structures of Computer Science Lab 6 Vectors and Matrices This lab consists of exercises on real-valued vectors and matrices. Most of the exercises will required pencil and paper. Put
More informationAn ecological approach to the classification of transient underwater acoustic events: Perceptual experiments and auditory models
An ecological approach to the classification of transient underwater acoustic events: Perceptual experiments and auditory models Simon Tucker Department of Computer Science University of Sheffield November
More informationMAT120 Supplementary Lecture Notes
MAT0 Supplementary Lecture Notes Overview of Numbers Integers: {..., 3,,, 0,,, 3,...} Rational Numbers:, 3, etc., quotients of integers a b, b 0; finite or repeating decimal expansions. Irrational Numbers:,,
More informationSparseness Constraints on Nonnegative Tensor Decomposition
Sparseness Constraints on Nonnegative Tensor Decomposition Na Li nali@clarksonedu Carmeliza Navasca cnavasca@clarksonedu Department of Mathematics Clarkson University Potsdam, New York 3699, USA Department
More informationTime-domain representations
Time-domain representations Speech Processing Tom Bäckström Aalto University Fall 2016 Basics of Signal Processing in the Time-domain Time-domain signals Before we can describe speech signals or modelling
More informationMultimedia Networking ECE 599
Multimedia Networking ECE 599 Prof. Thinh Nguyen School of Electrical Engineering and Computer Science Based on lectures from B. Lee, B. Girod, and A. Mukherjee 1 Outline Digital Signal Representation
More informationGeometric View of Machine Learning Nearest Neighbor Classification. Slides adapted from Prof. Carpuat
Geometric View of Machine Learning Nearest Neighbor Classification Slides adapted from Prof. Carpuat What we know so far Decision Trees What is a decision tree, and how to induce it from data Fundamental
More informationSingular value decomposition (SVD) of large random matrices. India, 2010
Singular value decomposition (SVD) of large random matrices Marianna Bolla Budapest University of Technology and Economics marib@math.bme.hu India, 2010 Motivation New challenge of multivariate statistics:
More informationCS123 INTRODUCTION TO COMPUTER GRAPHICS. Linear Algebra /34
Linear Algebra /34 Vectors A vector is a magnitude and a direction Magnitude = v Direction Also known as norm, length Represented by unit vectors (vectors with a length of 1 that point along distinct axes)
More informationUnsupervised learning: beyond simple clustering and PCA
Unsupervised learning: beyond simple clustering and PCA Liza Rebrova Self organizing maps (SOM) Goal: approximate data points in R p by a low-dimensional manifold Unlike PCA, the manifold does not have
More informationVoice Activity Detection Using Pitch Feature
Voice Activity Detection Using Pitch Feature Presented by: Shay Perera 1 CONTENTS Introduction Related work Proposed Improvement References Questions 2 PROBLEM speech Non speech Speech Region Non Speech
More informationMATH.2720 Introduction to Programming with MATLAB Vector and Matrix Algebra
MATH.2720 Introduction to Programming with MATLAB Vector and Matrix Algebra A. Vectors A vector is a quantity that has both magnitude and direction, like velocity. The location of a vector is irrelevant;
More informationFourier Analysis of Signals
Chapter 2 Fourier Analysis of Signals As we have seen in the last chapter, music signals are generally complex sound mixtures that consist of a multitude of different sound components. Because of this
More informationCMSC 422 Introduction to Machine Learning Lecture 4 Geometry and Nearest Neighbors. Furong Huang /
CMSC 422 Introduction to Machine Learning Lecture 4 Geometry and Nearest Neighbors Furong Huang / furongh@cs.umd.edu What we know so far Decision Trees What is a decision tree, and how to induce it from
More informationData Mining and Analysis
978--5-766- - Data Mining and Analysis: Fundamental Concepts and Algorithms CHAPTER Data Mining and Analysis Data mining is the process of discovering insightful, interesting, and novel patterns, as well
More informationCEPSTRAL ANALYSIS SYNTHESIS ON THE MEL FREQUENCY SCALE, AND AN ADAPTATIVE ALGORITHM FOR IT.
CEPSTRAL ANALYSIS SYNTHESIS ON THE EL FREQUENCY SCALE, AND AN ADAPTATIVE ALGORITH FOR IT. Summarized overview of the IEEE-publicated papers Cepstral analysis synthesis on the mel frequency scale by Satochi
More informationCognitive semantics and cognitive theories of representation: Session 6: Conceptual spaces
Cognitive semantics and cognitive theories of representation: Session 6: Conceptual spaces Martin Takáč Centre for cognitive science DAI FMFI Comenius University in Bratislava Príprava štúdia matematiky
More informationNon-negative Matrix Factorization: Algorithms, Extensions and Applications
Non-negative Matrix Factorization: Algorithms, Extensions and Applications Emmanouil Benetos www.soi.city.ac.uk/ sbbj660/ March 2013 Emmanouil Benetos Non-negative Matrix Factorization March 2013 1 / 25
More informationFOURIER ANALYSIS. (a) Fourier Series
(a) Fourier Series FOURIER ANAYSIS (b) Fourier Transforms Useful books: 1. Advanced Mathematics for Engineers and Scientists, Schaum s Outline Series, M. R. Spiegel - The course text. We follow their notation
More informationA Categorization of Mexican Free-Tailed Bat (Tadarida brasiliensis) Chirps
A Categorization of Mexican Free-Tailed Bat (Tadarida brasiliensis) Chirps ýýýýý Gregory Backus August 20, 2010 Abstract Male Mexican Free-tailed Bats (Tadarida brasiliensis) attract mates and defend territory
More informationReal Time Face Detection and Recognition using Haar - Based Cascade Classifier and Principal Component Analysis
Real Time Face Detection and Recognition using Haar - Based Cascade Classifier and Principal Component Analysis Sarala A. Dabhade PG student M. Tech (Computer Egg) BVDU s COE Pune Prof. Mrunal S. Bewoor
More informationESE 250: Digital Audio Basics. Week 4 February 5, The Frequency Domain. ESE Spring'13 DeHon, Kod, Kadric, Wilson-Shah
ESE 250: Digital Audio Basics Week 4 February 5, 2013 The Frequency Domain 1 Course Map 2 Musical Representation With this compact notation Could communicate a sound to pianist Much more compact than 44KHz
More informationPrediction of Dissimilarity Judgments between Tonal Sequences using Information Theory
Prediction of Dissimilarity Judgments between Tonal Sequences using Information Theory Michael Frishkopf University of Alberta 382 FAB, Department of Music Edmonton, Alberta, Canada T6G2C9 1-780-492-0225
More informationLinear Algebra & Geometry why is linear algebra useful in computer vision?
Linear Algebra & Geometry why is linear algebra useful in computer vision? References: -Any book on linear algebra! -[HZ] chapters 2, 4 Some of the slides in this lecture are courtesy to Prof. Octavia
More informationEnvironmental Cognition and Perception I
Environmental Cognition and Perception I Review: Spatial Interaction and Spatial Behavior II - Individual travel behavior - Activity space - Mental maps How we perceive the environment Maps in the head
More informationQuantum Mechanics for Scientists and Engineers. David Miller
Quantum Mechanics for Scientists and Engineers David Miller Vector spaces, operators and matrices Vector spaces, operators and matrices Vector space Vector space We need a space in which our vectors exist
More informationarxiv: v1 [cs.mm] 16 Feb 2016
Perceptual Vector Quantization for Video Coding Jean-Marc Valin and Timothy B. Terriberry Mozilla, Mountain View, USA Xiph.Org Foundation arxiv:1602.05209v1 [cs.mm] 16 Feb 2016 ABSTRACT This paper applies
More informationDesigning Information Devices and Systems I Fall 2015 Anant Sahai, Ali Niknejad Homework 2. This homework is due September 14, 2015, at Noon.
EECS 16A Designing Information Devices and Systems I Fall 2015 Anant Sahai, Ali Niknejad Homework 2 This homework is due September 14, 2015, at Noon. Submission Format Your homework submission should consist
More informationMTAEA Vectors in Euclidean Spaces
School of Economics, Australian National University January 25, 2010 Vectors. Economists usually work in the vector space R n. A point in this space is called a vector, and is typically defined by its
More informationUniversity of Florida CISE department Gator Engineering. Clustering Part 1
Clustering Part 1 Dr. Sanjay Ranka Professor Computer and Information Science and Engineering University of Florida, Gainesville What is Cluster Analysis? Finding groups of objects such that the objects
More informationGeometric and algebraic structures in pattern recognition
Geometric and algebraic structures in pattern recognition Luke Oeding Department of Mathematics, University of California, Berkeley April 30, 2012 Multimedia Pattern Recognition Rolf Bardeli mmprec.iais.fraunhofer.de/
More informationLINGUIST 716 Week 9: Compuational methods for finding dimensions
LINGUIST 716 Week 9: Compuational methods for finding dimensions Kristine Yu Department of Linguistics, UMass Amherst November 1, 2013 Computational methods for finding dimensions 716 Fall 2013 Week 9
More informationLecture 14 1/38 Phys 220. Final Exam. Wednesday, August 6 th 10:30 am 12:30 pm Phys multiple choice problems (15 points each 300 total)
Lecture 14 1/38 Phys 220 Final Exam Wednesday, August 6 th 10:30 am 12:30 pm Phys 114 20 multiple choice problems (15 points each 300 total) 75% will be from Chapters 10-16 25% from Chapters 1-9 Students
More informationMULTISCALE SCATTERING FOR AUDIO CLASSIFICATION
MULTISCALE SCATTERING FOR AUDIO CLASSIFICATION Joakim Andén CMAP, Ecole Polytechnique, 91128 Palaiseau anden@cmappolytechniquefr Stéphane Mallat CMAP, Ecole Polytechnique, 91128 Palaiseau ABSTRACT Mel-frequency
More informationCorrespondence Analysis & Related Methods
Correspondence Analysis & Related Methods Michael Greenacre SESSION 9: CA applied to rankings, preferences & paired comparisons Correspondence analysis (CA) can also be applied to other types of data:
More informationCSE 126 Multimedia Systems Midterm Exam (Form A)
University of California, San Diego Inst: Prof P. V. Rangan CSE 126 Multimedia Systems Midterm Exam (Form A) Spring 2003 Solution Assume the following input (before encoding) frame sequence (note that
More informationSOME SCALES THAT ARE SIMILAR TO THE CHROMATIC SCALE
SOME SCALES THAT ARE SIMILAR TO THE CHROMATIC SCALE WILL TURNER Abstract. We construct musical compositions by similarity, by constructing scales that are formally similar to the chromatic scale, and defining
More informationInteraction Analysis of Spatial Point Patterns
Interaction Analysis of Spatial Point Patterns Geog 2C Introduction to Spatial Data Analysis Phaedon C Kyriakidis wwwgeogucsbedu/ phaedon Department of Geography University of California Santa Barbara
More informationCS123 INTRODUCTION TO COMPUTER GRAPHICS. Linear Algebra 1/33
Linear Algebra 1/33 Vectors A vector is a magnitude and a direction Magnitude = v Direction Also known as norm, length Represented by unit vectors (vectors with a length of 1 that point along distinct
More informationVectors and their uses
Vectors and their uses Sharon Goldwater Institute for Language, Cognition and Computation School of Informatics, University of Edinburgh DRAFT Version 0.95: 3 Sep 2015. Do not redistribute without permission.
More informationFourier Analysis of Signals
Chapter 2 Fourier Analysis of Signals As we have seen in the last chapter, music signals are generally complex sound mixtures that consist of a multitude of different sound components. Because of this
More informationEXAMINING MUSICAL MEANING IN SIMILARITY THRESHOLDS
EXAMINING MUSICAL MEANING IN SIMILARITY THRESHOLDS Katherine M. Kinnaird Brown University katherine kinnaird@brown.edu ABSTRACT Many approaches to Music Information Retrieval tasks rely on correctly determining
More informationEstimation of Relative Operating Characteristics of Text Independent Speaker Verification
International Journal of Engineering Science Invention Volume 1 Issue 1 December. 2012 PP.18-23 Estimation of Relative Operating Characteristics of Text Independent Speaker Verification Palivela Hema 1,
More informationDesigning Information Devices and Systems I Spring 2018 Homework 11
EECS 6A Designing Information Devices and Systems I Spring 28 Homework This homework is due April 8, 28, at 23:59. Self-grades are due April 2, 28, at 23:59. Submission Format Your homework submission
More informationThe effect of speaking rate and vowel context on the perception of consonants. in babble noise
The effect of speaking rate and vowel context on the perception of consonants in babble noise Anirudh Raju Department of Electrical Engineering, University of California, Los Angeles, California, USA anirudh90@ucla.edu
More informationProc. of NCC 2010, Chennai, India
Proc. of NCC 2010, Chennai, India Trajectory and surface modeling of LSF for low rate speech coding M. Deepak and Preeti Rao Department of Electrical Engineering Indian Institute of Technology, Bombay
More informationTHE task of identifying the environment in which a sound
1 Feature Learning with Matrix Factorization Applied to Acoustic Scene Classification Victor Bisot, Romain Serizel, Slim Essid, and Gaël Richard Abstract In this paper, we study the usefulness of various
More informationLet x be an approximate solution for Ax = b, e.g., obtained by Gaussian elimination. Let x denote the exact solution. Call. r := b A x.
ESTIMATION OF ERROR Let x be an approximate solution for Ax = b, e.g., obtained by Gaussian elimination. Let x denote the exact solution. Call the residual for x. Then r := b A x r = b A x = Ax A x = A
More informationOBJECT CODING OF HARMONIC SOUNDS USING SPARSE AND STRUCTURED REPRESENTATIONS
OBJECT CODING OF HARMONIC SOUNDS USING SPARSE AND STRUCTURED REPRESENTATIONS Grégory Cornuz 1, Emmanuel Ravelli 1,2, 1 Institut Jean Le Rond d Alembert, LAM team Université Pierre et Marie Curie - Paris
More informationMULTIPITCH ESTIMATION AND INSTRUMENT RECOGNITION BY EXEMPLAR-BASED SPARSE REPRESENTATION. Ikuo Degawa, Kei Sato, Masaaki Ikehara
MULTIPITCH ESTIMATION AND INSTRUMENT RECOGNITION BY EXEMPLAR-BASED SPARSE REPRESENTATION Ikuo Degawa, Kei Sato, Masaaki Ikehara EEE Dept. Keio University Yokohama, Kanagawa 223-8522 Japan E-mail:{degawa,
More information22m:033 Notes: 6.1 Inner Product, Length and Orthogonality
m:033 Notes: 6. Inner Product, Length and Orthogonality Dennis Roseman University of Iowa Iowa City, IA http://www.math.uiowa.edu/ roseman April, 00 The inner product Arithmetic is based on addition and
More information