Automatic Speech Recognition handout (1)
|
|
- Blaise Bradley
- 5 years ago
- Views:
Transcription
1 Automatic Speech Recognition handout (1) Jan - Mar 2012 Revision : 1.1 Speech Signal Processing and Feature Extraction Hiroshi Shimodaira (h.shimodaira@ed.ac.uk)
2 Speech Communication Intention Language Motion Control Articulate organ (vocal tract) Understanding Language Auditory processing Auditory organs Signal source (vocal cords) speech sound Speaker Listener ASR (H. Shimodaira) I : 1
3 Spectrogram Waveform 8.0 Spectrogram Cross-section of spectrogram Frequency [khz] Time [s] 80 Intensity (db) Frequency (khz) ASR (H. Shimodaira) I : 2
4 Speech Production Model F1 F2 nasal cavity X( Ω) F3 ( 6dB/oct.) Nasal Cavity lips teeth oral cavity tongue pharynx H( Ω) lips x(t) +6dB/oct. F1 F2 Mouth Cavity (formants) F3 Larynx + Pharynx lungs larynx vocal folds Vocal Organs & Vocal Tract V( Ω) F0 =1/T Time domain: x(t) = h(t) v(t) = Fourier transform 12dB/oct. frequency 0 0 Ω vocal folds T 0 v(t) h(τ)v(t τ)dτ Frequency domain: X(Ω) = H(Ω)V (Ω) Ω : angular frequency (= 2πF ) F : frequency t ASR (H. Shimodaira) I : 3
5 Automatic Speech Recognition Find the word sequence W such that max W P (W X) = max W P (X W )P (W ) P (X) ASR (H. Shimodaira) I : 4
6 Signal Analysis for ASR Front-end analysis Convert acoustic signal into a sequence of feature vectors e.g. MFCCs, PLP cepstral coefficients x (t) c LPF (low pass filter) A/D conversion Sampling frequency F s Pre emphasis x[n] Spectral analysis Feature extraction Analysis window Frame shift c [k] m m: frame number k: feature index ASR (H. Shimodaira) I : 5
7 Feature parameters for ASR Features should contain sufficient information to distinguish phonemes / phones good time-resolutions [e.g. 10ms] good frequency-resolutions [e.g. 20 channels/bark-scale] not contain (or be separated from) F 0 and its harmonics be robust against speaker variation be robust against noise / channel distortions have good characteristics in terms of pattern recognition The number of features is as few as possible Features are independent of each other ASR (H. Shimodaira) I : 6
8 Converting analogue signals to machine readable form Discretisation (sampling) x c (t) x[n] continuous time discrete time continuous amplitude discrete amplitude Problem: information can be lost by sampling ASR (H. Shimodaira) I : 7
9 Sampling of continuous-time signals Continuous-time signal: x c (t) Modulated signal by a periodic impulse train: x s (t) = x c (t) δ(t nt s ) = n= n= x c (nt s )δ(t nt s ) Sampled signal: x[n] = x s (nt s ) discrete-time signal T s : Sampling interval ASR (H. Shimodaira) I : 8
10 Sampling of continuous-time signals(cont. 2) Q: Is the C/D conversion invertible? x c (t) C/D x[n] D/C x c (t)? ASR (H. Shimodaira) I : 9
11 Sampling of continuous-time signals(cont. 3) Q: Is the C/D conversion invertible? x c (t) C/D x[n] D/C x c (t)? A: No in general, but Yes under a special condition: Nyquist sampling theorem If x c (t) is band-limited (i.e. no frequency components > F s /2), then x c (t) can be fully reconstructed by x[n]. x c (t) = h Ts (t) k= h Ts (t) = sinc(t/t s ) = sin(πt/t s) πt/t s F s /2 : Nyquist Frequency, x[k]δ(t kt s ) = k= x[k]h Ts (t kt s ) F s = 1/T s : Sampling Frequency ASR (H. Shimodaira) I : 10
12 Sampling of continuous-time signals(cont. 4) Interpretation in frequency domain: X s (Ω) }{{} = Spectrum of x s (t) 1 T s k= Xc(Ω kω s ) }{{} Spectrum of x c (t) ASR (H. Shimodaira) I : 11
13 Sampling of continuous-time signals(cont. 5) x (t) c (low pass filter) Questions LPF A/D conversion Sampling frequency F s Pre emphasis x[n] Spectral analysis Feature extraction Analysis window Frame shift c [k] m m: frame number k: feature index 1. What sampling frequencies (F s ) are used for ASR? microphone voice: 12kHz 20kHz telephone voice: 8kHz 2. What are the advantages / disadvantages of using higher F s? 3. Why is pre-emphasis (+6dB/oct.) employed? x[n] = x 0 [n] ax 0 [n 1], a = ASR (H. Shimodaira) I : 12
14 Spectral analysis: Fourier Transform FT for continuous-time signals (& continuous-frequency) X c (Ω) = x c (t)e jωt dt x c (t) = 1 2π X c (Ω)e jωt dω (time domain freq. domain) (freq. domain time domain) FT for discrete-time signals (& continuous-frequency) X(e jω ) = x[n] = 1 2π n= π π x[n]e jωn X(e jω )e jωn dω X(e jω ) 2 Power spectrum log X(e jω ) 2 Log power spectrum where ω = T s Ω = 2πf, e jωn = cos(ωn) + j sin(ωn), j : the imaginary unit ASR (H. Shimodaira) I : 13
15 An interpretation of FT Inner product between two vectors (Linear Algebra) 2-dimensional case a = (a 1, a 2 ) t a b = (b 1, b 2 ) t a b = a t b = a 1 b 1 + a 2 b 2 = a b cos θ b b = Infinite-dimensional case θ a cos θ x {x[n]} e ω { e jωn} = {cos(ωn) + j sin(ωn)} cos ω + jsin ω X(e jω ) = n= if 1 x[n]e jωn = x e jωn = x cos ω + jx sin ω x cos ω : proportion of how much cos ω component is contained in x ASR (H. Shimodaira) I : 14
16 Short-time Spectrum Analysis Problem with FT Assuming signals are stationary: signal properties do not change over time If signals are non-stationary loses information on time varying features Short-time Fourier transform (STFT) (Time-dependent Fourier transform) Divide the signal x[n] into short-time segments (frames) x k [m] and apply FT to each segment. x[n] x 1 [m], x 2 [m],..., x k [m],... X(ω) X 1 (ω), X 2 (ω),..., X k (ω),... ASR (H. Shimodaira) I : 15
17 Short-time Spectrum Analysis(cont. 2) windowing shift frame 70 Intensity Discrete Fourier Transform Time (frame) 60 Frequency Short time power spectrum Frequency 0 10 ASR (H. Shimodaira) I : 16
18 Short-time Spectrum Analysis(cont. 3) Trade-off problem of short time spectrum analysis frequency resolution time resolution a compromise for ASR: window width short long window width (frame width): ms window shift (frame shift): 5 15 ms ASR (H. Shimodaira) I : 17
19 The Effect of Windowing in STFT Time domain: y k [n] = w k [n]x[n], w k [n] : time-window for k-th frame Simply cutting out a short segment (frame) from x[n] implies applying a rectangular window on to x[n]. causes discontinuities at the edges of the segment. Instead, a tapered window is usually used.. e.g. Hamming (α = ) or Hanning (α = 0.5) window) ( ) 2πl w[l] = (1 α) α cos N : window width N 1 1 rectangle 1 hammin 1 hannin 1 blackman 1 bartlett rectangle Hamming Hanning Blackman Bartlett ASR (H. Shimodaira) I : 18
20 The Effect of Windowing in STFT(cont. 2) Frequency domain: Y k (e jω ) = 1 2π π π W k (e jθ )X(e j(ω θ) )dθ Periodic convolution Power spectrum of the frame is given as a periodic convolution between the power spectra of x[n] and w k [n]. If we want Y k (e jω ) = X(e jω ), the necessary and sufficient condition for this is W k (e jω ) = δ(ω), i.e. w k [n] = F 1 δ(ω) = 1, which means the length of w k [n] is infinite. there is no window function of finite length that causes no distortion. NB: hereafter x[n] will be also used to denote a segmented signal for simplicity. ASR (H. Shimodaira) I : 19
21 The Effect of Windowing in STFT(cont. 3) Spectral analysis of two sine signals of close frequencies ASR (H. Shimodaira) I : 20
22 Problems with STFT The estimated power spectrum contains harmonics of F 0, which makes it difficult to estimate the envelope of the spectrum. Frequency bins of STFT are highly correlated each other, i.e. power spectrum representation is highly redundant Log X(w) ASR (H. Shimodaira) I : 21
23 Cepstrum Analysis Idea: split(deconvolve) the power spectrum into spectrum envelope and F 0 harmonics Log X(w) Cepstrum Log-spectrum [freq. domain] Inverse Fourier Transform Cepstrum [time domain] (quefrency) Liftering to get low/high part (lifter: filter used in cepstral domain) Fourier Transform Envelope (Lag=30) Residue Smoothed-spectrum [freq. domain] (low-part of cepstrum) Log-spectrum of high-part of cepstrum ASR (H. Shimodaira) I : 22
24 Cepstrum Analysis(cont. 2) Log spectrum h[n] : vocal tract x[n] = h[n] v[n] v[n] : glottal sounds F X(e jω ) = H(e jω )V (e jω ) log (Fourier transform) log X(e jω ) = log H(e jω ) }{{} + log V (ejω ) }{{} Cepstrum (spectral envelope) F 1 c(τ) = F 1 { log X(e jω ) } (spectral fine structure) = F 1 { log H(e jω ) } + F 1 { log V (e jω ) } ASR (H. Shimodaira) I : 23
25 LPC Analysis Linear Predictive Coding (LPC): a model-based / parametric spectrum estimation Assume a linear system for human speech production sound source v[n] vocal tract speech x[n] v[n] h[n] x[n] h[n] : impulse response x[n] = h[n] v[n] = k=0 h[k] v[n k] Using a model enables us to estimate a spectrum of vocal tract from small amount of observations represent the spectrum with a small number of parameters synthesise speech with the parameters ASR (H. Shimodaira) I : 24
26 LPC Analysis(cont. 2) Predict x[n] from x[n 1], x[n 2], ˆx[n] = N k=1 a k x[n k] e[n] = x[n] ˆx[n] = x[n] N k=1 a k x[n k] prediction error Optimisation problem Find {a k } that minimises the mean square (MS) error: P e = E { e 2 [n] } ( ) 2 N = E x[n] a k x[n k] k=1 {a k } : LPC coefficients ASR (H. Shimodaira) I : 25
27 Spectrums estimated by FT & LPC ASR (H. Shimodaira) I : 26
28 LPC summary Spectrum can be modelled/coded with around 14LP Cs. LPC family PARCOR (Partial Auto-Correlation Coefficient) LSP (Line Spectral Pairs) / LSF (Line Spectrum Frequencies) CSM (Composite Sinusoidal Model) LPC can be used to predict log-area ratio coefficients lossless tube model LPC-(Mel)Cepstrum: LPC based cepstrum. Drawback: LPC assumes AR model which does not suit to model nasal sounds that have zeros in spectrum. Difficult to determine the prediction order N. ASR (H. Shimodaira) I : 27
29 Taking into Perceptual Attributes Physical quality Intensity Fundamental frequency Spectral shape Onset/offset time Phase difference in binaural hearing Technical terms equal-loudness contours masking auditory filters (critical-band filters) critical bandwidth Perceptual quality Loudness Pitch Timbre Timing Location ASR (H. Shimodaira) I : 28
30 Taking into Perceptual Attributes(cont. 2) ASR (H. Shimodaira) I : 29
31 Taking into Perceptual Attributes(cont. 3) Non-linear frequency scale Bark scale b(f) = 13 arctan( f) arctan((f/7500) 2 ) [Bark] Mel scale B(f) = 1127 ln(1 + f/700) Bark frequency [Bark] linear frequency [Hz] warped normalized frequency linear frequency [Hz] ln Bark Mel ASR (H. Shimodaira) I : 30
32 Filter Bank Analysis Speech x[n] Bandpass Filter 1 Bandpass Filter K x [n] 1 x [n] K ω ω ω K ω x i [n] = h i [n] x[n] = M i 1 k=0 h i [k]x[n k] h i [n]: Impulse response of Bandpass filter i ω perceptual scale ASR (H. Shimodaira) I : 31
33 Filter Bank Analysis(cont. 2) Speech Bandpass Filter 1 x [n] 1 Nonlinearity v [n] 1 Lowpass Filter y [n] 1 Down Sampling x[n] Bandpass Filter K x [n] K Nonlinearity v [n] K Lowpass Filter y [n] K Down Sampling v 0 ω x 0 ω Trade-off problem Freq. resolution # of filters length of filter Time resolution ASR (H. Shimodaira) I : 32
34 Filter Bank Analysis(cont. 3) Another implementation of filter banks: apply a mel-scale filter bank to STFT power spectrum to obtain mel-scale power spectrum DFT(STFT) power spectrum Triangular band pass filters Frequency bins Mel scale power spectrum ASR (H. Shimodaira) I : 33
35 MFCC MFCC: Mel-frequency Cepstral Coefficients c[n] x[n] DFT X[k] X[k] 2 DCT: c[n] = 2 N N i=1 Mel-frequency filterbank ( ) πn(i 0.5) s[i] cos N log S[m] DCT c[n], where s[i] = log S[i] DFT: discrete Fourier transform, DCT: discrete cosine transform MFCCs are widely used in HMM-based ASR systems. The first 12 MFCCs (c[1] c[12]) are generally used. ASR (H. Shimodaira) I : 34
36 MFCC(cont. 2) MFCCs are less correlated each other than DCT/Filter-bank based spectrum. Good compression rate. Feature dimensionality / frame Speech wave 400 DCT Spectrum Filter-bank MFCC 12 where F s = 16kHz, frame-width = 25ms, frame-shift = 10ms are assumed. MFCCs show better ASR performance than filter-bank features, but MFCCs are not robust against noises. ASR (H. Shimodaira) I : 35
37 Perceptually-based Linear Prediction (PLP) [Hermansky, 1985,1990] PLP had been shown experimentally to be more noise robust more speaker independent than MFCCs ASR (H. Shimodaira) I : 36
38 Other features with low dimensionality Formants (F 1, F 2, F 3, ) They are not used in modern ASR systems, but why? ASR (H. Shimodaira) I : 37
39 Using temporal features: dynamic features In SP lab-sessions on speech recognition using HTK, MFCCs, and energy MFCCs, energy 2 MFCCs, 2 energy, 2 : delta features (dynamic features / time derivatives) [Furui, 1986] continuous time discrete time c(t) c[n] c (t) = dc(t) M c[n] w i c[n + i] e.g. c[n] = dt i= M c (t) = d2 c(t) 2 M c[n] w dt 2 i c[n + i] i= M c[n + 1] c[n 1] 2 ASR (H. Shimodaira) I : 38
40 Using temporal features: dynamic features(cont. 2) c(t) c (t ) 0 t 0 time ASR (H. Shimodaira) I : 39
41 Using temporal features: dynamic features(cont. 3) An acoustic feature vector, eg MFCCs, representing part of a speech signal is highly correlated with its neighbours. HMM based acoustic models assume there is no dependency between the observations. Those correlations can be captured to some extent by augmenting the original set of static acoustic features, eg. MFCCs, with dynamic features. ASR (H. Shimodaira) I : 40
42 General Feature Transformation Orthogonal transformation (orthogonal bases) DCT (discrete cosine transform) PCA (principal component analysis) Transformation based on the bases that maximises the separability between classes. LDA (linear discriminant analysis) / Fisher s linear discrminant HLDA (heteroscedastic linear discriminant analysis) ASR (H. Shimodaira) I : 41
43 A comparison of speech features I. Mporas, et al., Comparison of Speech Features on the Speech Recognition Task, Journal of Computer Science, Vol.3, pp , NB SBC WPSR OWPF WPSR LFCC-FB HFCC-FB Feature WER(%) SER(%) SBC (16) WPSR125 (16) OWPF (16) LFCC-FB HFCC-FB HFCC-FB PLP-FB MFCC-FB Subband-based Cepstral Coefficients Wavelet packet features Overlapping wavelet packet features Wavelet packet-based speech features Linear-spaced filter-bank based cepstral coefficients Human factor cepstral coefficients The above result was obtained for TIMIT speech corpus. Results might change a lot under different conditions (e.g. noise, tasks, ASR systems) ASR (H. Shimodaira) I : 42
44 Further topics on feature extraction Feature normalisation/enhancement in terms of noise / environments speakers / speaking styles speech recognition Pitch (F 0 ) adapted feature extraction ASR (H. Shimodaira) I : 43
45 SUMMARY Nyquist Sampling theory Short-time Spectrum Analysis Non-parametric method Short-time Fourier Transform Cepstrum, MFCC Filter bank Parametric methods LPC, PLP Windowing effect: trade-off between time and frequency resolutions Dynamic features (delta features) There is no best feature that can be used for any purposes, but MFCC is widely used for ASR and TTS. ASR (H. Shimodaira) I : 44
46 SUMMARY(cont. 2) Front-end analysis has a great influence on ASR performance. For robust ASR in real environments, various techniques for front-end processing have been proposed. e.g. spectral subtraction (SS), cepstral mean normalisation (CMN) Spectrum analysis and feature extraction involve information loss and non-linear distortions. There is always a tradeoff between accuracy and efficiency. (e.g. spatial resolution vs. temporal resolution) ASR (H. Shimodaira) I : 45
47 References John N. Holmes, Wendy J. Holmes, Speech Synthesis and Recognition, Taylor and Francis (2001), 2nd edition (chapter 2, 4, 10) ajr/speechanalysis/ B. Gold, N. Morgan, Speech and Audio Signal Processing: Processing and Perception of Speech and Music, John Wiley and Sons (1999). Spoken language processing: a guide to theory, algorithm, and system development, Xuedong Huang, Alex Acero and Hsiao-Wuen Hon, Prentice Hall (2001). isbn: ASR (H. Shimodaira) I : 46
48 References(cont. 2) Robusness in Automatic Speech Recognition, J-C Junqua and J-P Hanton,, Kluwer Academic Publications (1996). isbn: A Comparative Study of Traditional and Newly Proposed Features for Recognition of Speech Under Stress, Sahar Bou-Ghazale and John H.L. Hansen, IEEE Trans SAP, vol. 8, no. 4, pp , July ASR (H. Shimodaira) I : 47
Speech Production. Automatic Speech Recognition handout (1) Jan - Mar 2009 Revision : 1.1. Speech Communication. Spectrogram. Waveform.
Speech Production Automatic Speech Recognition handout () Jan - Mar 29 Revision :. Speech Signal Processing and Feature Extraction lips teeth nasal cavity oral cavity tongue lang S( Ω) pharynx larynx vocal
More informationSpeech Signal Analysis
Speech Signal Analysis Hiroshi Shimodaira and Steve Renals Automatic Speech Recognition ASR Lectures 2&3 14,18 January 216 ASR Lectures 2&3 Speech Signal Analysis 1 Overview Speech Signal Analysis for
More informationSignal Processing for Speech Applications - Part 2-1. Signal Processing For Speech Applications - Part 2
Signal Processing for Speech Applications - Part 2-1 Signal Processing For Speech Applications - Part 2 May 14, 2013 Signal Processing for Speech Applications - Part 2-2 References Huang et al., Chapter
More informationCepstrum alanysis of speech signals
Cepstrum alanysis of speech signals ELEC-E5520 Speech and language processing methods Spring 2016 Mikko Kurimo 1 /48 Contents Literature and other material Idea and history of cepstrum Cepstrum and LP
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 informationspeech signal S(n). This involves a transformation of S(n) into another signal or a set of signals
16 3. SPEECH ANALYSIS 3.1 INTRODUCTION TO SPEECH ANALYSIS Many speech processing [22] applications exploits speech production and perception to accomplish speech analysis. By speech analysis we extract
More informationSpeech Processing. Undergraduate course code: LASC10061 Postgraduate course code: LASC11065
Speech Processing Undergraduate course code: LASC10061 Postgraduate course code: LASC11065 All course materials and handouts are the same for both versions. Differences: credits (20 for UG, 10 for PG);
More informationAdvanced audio analysis. Martin Gasser
Advanced audio analysis Martin Gasser Motivation Which methods are common in MIR research? How can we parameterize audio signals? Interesting dimensions of audio: Spectral/ time/melody structure, high
More informationTopic. Spectrogram Chromagram Cesptrogram. Bryan Pardo, 2008, Northwestern University EECS 352: Machine Perception of Music and Audio
Topic Spectrogram Chromagram Cesptrogram Short time Fourier Transform Break signal into windows Calculate DFT of each window The Spectrogram spectrogram(y,1024,512,1024,fs,'yaxis'); A series of short term
More informationI D I A P. On Factorizing Spectral Dynamics for Robust Speech Recognition R E S E A R C H R E P O R T. Iain McCowan a Hemant Misra a,b
R E S E A R C H R E P O R T I D I A P On Factorizing Spectral Dynamics for Robust Speech Recognition a Vivek Tyagi Hervé Bourlard a,b IDIAP RR 3-33 June 23 Iain McCowan a Hemant Misra a,b to appear in
More informationMel Spectrum Analysis of Speech Recognition using Single Microphone
International Journal of Engineering Research in Electronics and Communication Mel Spectrum Analysis of Speech Recognition using Single Microphone [1] Lakshmi S.A, [2] Cholavendan M [1] PG Scholar, Sree
More informationI D I A P. Mel-Cepstrum Modulation Spectrum (MCMS) Features for Robust ASR R E S E A R C H R E P O R T. Iain McCowan a Hemant Misra a,b
R E S E A R C H R E P O R T I D I A P Mel-Cepstrum Modulation Spectrum (MCMS) Features for Robust ASR a Vivek Tyagi Hervé Bourlard a,b IDIAP RR 3-47 September 23 Iain McCowan a Hemant Misra a,b to appear
More informationInternational Journal of Engineering and Techniques - Volume 1 Issue 6, Nov Dec 2015
RESEARCH ARTICLE OPEN ACCESS A Comparative Study on Feature Extraction Technique for Isolated Word Speech Recognition Easwari.N 1, Ponmuthuramalingam.P 2 1,2 (PG & Research Department of Computer Science,
More informationSpeech Synthesis; Pitch Detection and Vocoders
Speech Synthesis; Pitch Detection and Vocoders Tai-Shih Chi ( 冀泰石 ) Department of Communication Engineering National Chiao Tung University May. 29, 2008 Speech Synthesis Basic components of the text-to-speech
More informationAuditory Based Feature Vectors for Speech Recognition Systems
Auditory Based Feature Vectors for Speech Recognition Systems Dr. Waleed H. Abdulla Electrical & Computer Engineering Department The University of Auckland, New Zealand [w.abdulla@auckland.ac.nz] 1 Outlines
More informationE : Lecture 8 Source-Filter Processing. E : Lecture 8 Source-Filter Processing / 21
E85.267: Lecture 8 Source-Filter Processing E85.267: Lecture 8 Source-Filter Processing 21-4-1 1 / 21 Source-filter analysis/synthesis n f Spectral envelope Spectral envelope Analysis Source signal n 1
More informationSAMPLING THEORY. Representing continuous signals with discrete numbers
SAMPLING THEORY Representing continuous signals with discrete numbers Roger B. Dannenberg Professor of Computer Science, Art, and Music Carnegie Mellon University ICM Week 3 Copyright 2002-2013 by Roger
More informationURBANA-CHAMPAIGN. CS 498PS Audio Computing Lab. Audio DSP basics. Paris Smaragdis. paris.cs.illinois.
UNIVERSITY ILLINOIS @ URBANA-CHAMPAIGN OF CS 498PS Audio Computing Lab Audio DSP basics Paris Smaragdis paris@illinois.edu paris.cs.illinois.edu Overview Basics of digital audio Signal representations
More informationDigital Signal Processing
Digital Signal Processing Lecture 9 Discrete-Time Processing of Continuous-Time Signals Alp Ertürk alp.erturk@kocaeli.edu.tr Analog to Digital Conversion Most real life signals are analog signals These
More informationLinguistic Phonetics. Spectral Analysis
24.963 Linguistic Phonetics Spectral Analysis 4 4 Frequency (Hz) 1 Reading for next week: Liljencrants & Lindblom 1972. Assignment: Lip-rounding assignment, due 1/15. 2 Spectral analysis techniques There
More informationUniversity of Washington Department of Electrical Engineering Computer Speech Processing EE516 Winter 2005
University of Washington Department of Electrical Engineering Computer Speech Processing EE516 Winter 2005 Lecture 5 Slides Jan 26 th, 2005 Outline of Today s Lecture Announcements Filter-bank analysis
More informationIsolated Digit Recognition Using MFCC AND DTW
MarutiLimkar a, RamaRao b & VidyaSagvekar c a Terna collegeof Engineering, Department of Electronics Engineering, Mumbai University, India b Vidyalankar Institute of Technology, Department ofelectronics
More informationIntroduction of Audio and Music
1 Introduction of Audio and Music Wei-Ta Chu 2009/12/3 Outline 2 Introduction of Audio Signals Introduction of Music 3 Introduction of Audio Signals Wei-Ta Chu 2009/12/3 Li and Drew, Fundamentals of Multimedia,
More informationEE482: Digital Signal Processing Applications
Professor Brendan Morris, SEB 3216, brendan.morris@unlv.edu EE482: Digital Signal Processing Applications Spring 2014 TTh 14:30-15:45 CBC C222 Lecture 14 Quiz 04 Review 14/04/07 http://www.ee.unlv.edu/~b1morris/ee482/
More informationLecture 6: Speech modeling and synthesis
EE E682: Speech & Audio Processing & Recognition Lecture 6: Speech modeling and synthesis 1 2 3 4 5 Modeling speech signals Spectral and cepstral models Linear Predictive models (LPC) Other signal models
More informationAN ANALYSIS OF SPEECH RECOGNITION PERFORMANCE BASED UPON NETWORK LAYERS AND TRANSFER FUNCTIONS
AN ANALYSIS OF SPEECH RECOGNITION PERFORMANCE BASED UPON NETWORK LAYERS AND TRANSFER FUNCTIONS Kuldeep Kumar 1, R. K. Aggarwal 1 and Ankita Jain 2 1 Department of Computer Engineering, National Institute
More informationPerformance Analysis of MFCC and LPCC Techniques in Automatic Speech Recognition
www.ijecs.in International Journal Of Engineering And Computer Science ISSN:2319-7242 Volume - 3 Issue - 8 August, 2014 Page No. 7727-7732 Performance Analysis of MFCC and LPCC Techniques in Automatic
More informationTopic 2. Signal Processing Review. (Some slides are adapted from Bryan Pardo s course slides on Machine Perception of Music)
Topic 2 Signal Processing Review (Some slides are adapted from Bryan Pardo s course slides on Machine Perception of Music) Recording Sound Mechanical Vibration Pressure Waves Motion->Voltage Transducer
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 informationPreeti Rao 2 nd CompMusicWorkshop, Istanbul 2012
Preeti Rao 2 nd CompMusicWorkshop, Istanbul 2012 o Music signal characteristics o Perceptual attributes and acoustic properties o Signal representations for pitch detection o STFT o Sinusoidal model o
More informationLecture 5: Speech modeling. The speech signal
EE E68: Speech & Audio Processing & Recognition Lecture 5: Speech modeling 1 3 4 5 Modeling speech signals Spectral and cepstral models Linear Predictive models (LPC) Other signal models Speech synthesis
More informationDERIVATION OF TRAPS IN AUDITORY DOMAIN
DERIVATION OF TRAPS IN AUDITORY DOMAIN Petr Motlíček, Doctoral Degree Programme (4) Dept. of Computer Graphics and Multimedia, FIT, BUT E-mail: motlicek@fit.vutbr.cz Supervised by: Dr. Jan Černocký, Prof.
More informationEC 6501 DIGITAL COMMUNICATION UNIT - II PART A
EC 6501 DIGITAL COMMUNICATION 1.What is the need of prediction filtering? UNIT - II PART A [N/D-16] Prediction filtering is used mostly in audio signal processing and speech processing for representing
More informationQuantification of glottal and voiced speech harmonicsto-noise ratios using cepstral-based estimation
Quantification of glottal and voiced speech harmonicsto-noise ratios using cepstral-based estimation Peter J. Murphy and Olatunji O. Akande, Department of Electronic and Computer Engineering University
More informationEE482: Digital Signal Processing Applications
Professor Brendan Morris, SEB 3216, brendan.morris@unlv.edu EE482: Digital Signal Processing Applications Spring 2014 TTh 14:30-15:45 CBC C222 Lecture 12 Speech Signal Processing 14/03/25 http://www.ee.unlv.edu/~b1morris/ee482/
More informationSPEech Feature Toolbox (SPEFT) Design and Emotional Speech Feature Extraction
SPEech Feature Toolbox (SPEFT) Design and Emotional Speech Feature Extraction by Xi Li A thesis submitted to the Faculty of Graduate School, Marquette University, in Partial Fulfillment of the Requirements
More informationCOMP 546, Winter 2017 lecture 20 - sound 2
Today we will examine two types of sounds that are of great interest: music and speech. We will see how a frequency domain analysis is fundamental to both. Musical sounds Let s begin by briefly considering
More informationRASTA-PLP SPEECH ANALYSIS. Aruna Bayya. Phil Kohn y TR December 1991
RASTA-PLP SPEECH ANALYSIS Hynek Hermansky Nelson Morgan y Aruna Bayya Phil Kohn y TR-91-069 December 1991 Abstract Most speech parameter estimation techniques are easily inuenced by the frequency response
More informationComparison of Spectral Analysis Methods for Automatic Speech Recognition
INTERSPEECH 2013 Comparison of Spectral Analysis Methods for Automatic Speech Recognition Venkata Neelima Parinam, Chandra Vootkuri, Stephen A. Zahorian Department of Electrical and Computer Engineering
More informationChapter 7. Frequency-Domain Representations 语音信号的频域表征
Chapter 7 Frequency-Domain Representations 语音信号的频域表征 1 General Discrete-Time Model of Speech Production Voiced Speech: A V P(z)G(z)V(z)R(z) Unvoiced Speech: A N N(z)V(z)R(z) 2 DTFT and DFT of Speech The
More informationCS3291: Digital Signal Processing
CS39 Exam Jan 005 //08 /BMGC University of Manchester Department of Computer Science First Semester Year 3 Examination Paper CS39: Digital Signal Processing Date of Examination: January 005 Answer THREE
More informationT Automatic Speech Recognition: From Theory to Practice
Automatic Speech Recognition: From Theory to Practice http://www.cis.hut.fi/opinnot// September 27, 2004 Prof. Bryan Pellom Department of Computer Science Center for Spoken Language Research University
More informationDigital Signal Processing
COMP ENG 4TL4: Digital Signal Processing Notes for Lecture #27 Tuesday, November 11, 23 6. SPECTRAL ANALYSIS AND ESTIMATION 6.1 Introduction to Spectral Analysis and Estimation The discrete-time Fourier
More informationProject 0: Part 2 A second hands-on lab on Speech Processing Frequency-domain processing
Project : Part 2 A second hands-on lab on Speech Processing Frequency-domain processing February 24, 217 During this lab, you will have a first contact on frequency domain analysis of speech signals. You
More informationEE123 Digital Signal Processing
EE123 Digital Signal Processing Lecture 5A Time-Frequency Tiling Subtleties in filtering/processing with DFT x[n] H(e j! ) y[n] System is implemented by overlap-and-save Filtering using DFT H[k] π 2π Subtleties
More informationRobust Algorithms For Speech Reconstruction On Mobile Devices
Robust Algorithms For Speech Reconstruction On Mobile Devices XU SHAO A Thesis presented for the degree of Doctor of Philosophy Speech Group School of Computing Sciences University of East Anglia England
More informationSynthesis Techniques. Juan P Bello
Synthesis Techniques Juan P Bello Synthesis It implies the artificial construction of a complex body by combining its elements. Complex body: acoustic signal (sound) Elements: parameters and/or basic signals
More informationSingle Channel Speaker Segregation using Sinusoidal Residual Modeling
NCC 2009, January 16-18, IIT Guwahati 294 Single Channel Speaker Segregation using Sinusoidal Residual Modeling Rajesh M Hegde and A. Srinivas Dept. of Electrical Engineering Indian Institute of Technology
More informationSpeech and Music Discrimination based on Signal Modulation Spectrum.
Speech and Music Discrimination based on Signal Modulation Spectrum. Pavel Balabko June 24, 1999 1 Introduction. This work is devoted to the problem of automatic speech and music discrimination. As we
More informationIntroducing COVAREP: A collaborative voice analysis repository for speech technologies
Introducing COVAREP: A collaborative voice analysis repository for speech technologies John Kane Wednesday November 27th, 2013 SIGMEDIA-group TCD COVAREP - Open-source speech processing repository 1 Introduction
More information8.3 Basic Parameters for Audio
8.3 Basic Parameters for Audio Analysis Physical audio signal: simple one-dimensional amplitude = loudness frequency = pitch Psycho-acoustic features: complex A real-life tone arises from a complex superposition
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 informationAnalysis/synthesis coding
TSBK06 speech coding p.1/32 Analysis/synthesis coding Many speech coders are based on a principle called analysis/synthesis coding. Instead of coding a waveform, as is normally done in general audio coders
More informationClassification of ships using autocorrelation technique for feature extraction of the underwater acoustic noise
Classification of ships using autocorrelation technique for feature extraction of the underwater acoustic noise Noha KORANY 1 Alexandria University, Egypt ABSTRACT The paper applies spectral analysis to
More informationSOUND SOURCE RECOGNITION AND MODELING
SOUND SOURCE RECOGNITION AND MODELING CASA seminar, summer 2000 Antti Eronen antti.eronen@tut.fi Contents: Basics of human sound source recognition Timbre Voice recognition Recognition of environmental
More informationFilter Banks I. Prof. Dr. Gerald Schuller. Fraunhofer IDMT & Ilmenau University of Technology Ilmenau, Germany. Fraunhofer IDMT
Filter Banks I Prof. Dr. Gerald Schuller Fraunhofer IDMT & Ilmenau University of Technology Ilmenau, Germany 1 Structure of perceptual Audio Coders Encoder Decoder 2 Filter Banks essential element of most
More informationSpeech Enhancement in Presence of Noise using Spectral Subtraction and Wiener Filter
Speech Enhancement in Presence of Noise using Spectral Subtraction and Wiener Filter 1 Gupteswar Sahu, 2 D. Arun Kumar, 3 M. Bala Krishna and 4 Jami Venkata Suman Assistant Professor, Department of ECE,
More informationLaboratory Assignment 4. Fourier Sound Synthesis
Laboratory Assignment 4 Fourier Sound Synthesis PURPOSE This lab investigates how to use a computer to evaluate the Fourier series for periodic signals and to synthesize audio signals from Fourier series
More informationMel- frequency cepstral coefficients (MFCCs) and gammatone filter banks
SGN- 14006 Audio and Speech Processing Pasi PerQlä SGN- 14006 2015 Mel- frequency cepstral coefficients (MFCCs) and gammatone filter banks Slides for this lecture are based on those created by Katariina
More informationNCCF ACF. cepstrum coef. error signal > samples
ESTIMATION OF FUNDAMENTAL FREQUENCY IN SPEECH Petr Motl»cek 1 Abstract This paper presents an application of one method for improving fundamental frequency detection from the speech. The method is based
More informationAudio processing methods on marine mammal vocalizations
Audio processing methods on marine mammal vocalizations Xanadu Halkias Laboratory for the Recognition and Organization of Speech and Audio http://labrosa.ee.columbia.edu Sound to Signal sound is pressure
More informationFFT analysis in practice
FFT analysis in practice Perception & Multimedia Computing Lecture 13 Rebecca Fiebrink Lecturer, Department of Computing Goldsmiths, University of London 1 Last Week Review of complex numbers: rectangular
More informationPerceptually Motivated Linear Prediction Cepstral Features for Network Speech Recognition
Perceptually Motivated Linear Prediction Cepstral Features for Network Speech Recognition Aadel Alatwi, Stephen So, Kuldip K. Paliwal Signal Processing Laboratory Griffith University, Brisbane, QLD, 4111,
More informationComplex Sounds. Reading: Yost Ch. 4
Complex Sounds Reading: Yost Ch. 4 Natural Sounds Most sounds in our everyday lives are not simple sinusoidal sounds, but are complex sounds, consisting of a sum of many sinusoids. The amplitude and frequency
More informationDesign and Implementation of Speech Recognition Systems
Design and Implementation of Speech Recognition Systems Spring 2013 Class 3: Feature Computation 30 Jan 2013 1 First Step: Feature Extraction Speech recognition is a type of pattern recognition problem
More informationAn Improved Voice Activity Detection Based on Deep Belief Networks
e-issn 2455 1392 Volume 2 Issue 4, April 2016 pp. 676-683 Scientific Journal Impact Factor : 3.468 http://www.ijcter.com An Improved Voice Activity Detection Based on Deep Belief Networks Shabeeba T. K.
More informationPerformance study of Text-independent Speaker identification system using MFCC & IMFCC for Telephone and Microphone Speeches
Performance study of Text-independent Speaker identification system using & I for Telephone and Microphone Speeches Ruchi Chaudhary, National Technical Research Organization Abstract: A state-of-the-art
More informationCommunications Theory and Engineering
Communications Theory and Engineering Master's Degree in Electronic Engineering Sapienza University of Rome A.A. 2018-2019 Speech and telephone speech Based on a voice production model Parametric representation
More informationUniversity of Colorado at Boulder ECEN 4/5532. Lab 1 Lab report due on February 2, 2015
University of Colorado at Boulder ECEN 4/5532 Lab 1 Lab report due on February 2, 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 informationOutline. Introduction to Biosignal Processing. Overview of Signals. Measurement Systems. -Filtering -Acquisition Systems (Quantisation and Sampling)
Outline Overview of Signals Measurement Systems -Filtering -Acquisition Systems (Quantisation and Sampling) Digital Filtering Design Frequency Domain Characterisations - Fourier Analysis - Power Spectral
More informationDimension Reduction of the Modulation Spectrogram for Speaker Verification
Dimension Reduction of the Modulation Spectrogram for Speaker Verification Tomi Kinnunen Speech and Image Processing Unit Department of Computer Science University of Joensuu, Finland Kong Aik Lee and
More informationCG401 Advanced Signal Processing. Dr Stuart Lawson Room A330 Tel: January 2003
CG40 Advanced Dr Stuart Lawson Room A330 Tel: 23780 e-mail: ssl@eng.warwick.ac.uk 03 January 2003 Lecture : Overview INTRODUCTION What is a signal? An information-bearing quantity. Examples of -D and 2-D
More informationDiscrete Fourier Transform (DFT)
Amplitude Amplitude Discrete Fourier Transform (DFT) DFT transforms the time domain signal samples to the frequency domain components. DFT Signal Spectrum Time Frequency DFT is often used to do frequency
More informationSPEECH AND SPECTRAL ANALYSIS
SPEECH AND SPECTRAL ANALYSIS 1 Sound waves: production in general: acoustic interference vibration (carried by some propagation medium) variations in air pressure speech: actions of the articulatory organs
More informationECE 556 BASICS OF DIGITAL SPEECH PROCESSING. Assıst.Prof.Dr. Selma ÖZAYDIN Spring Term-2017 Lecture 2
ECE 556 BASICS OF DIGITAL SPEECH PROCESSING Assıst.Prof.Dr. Selma ÖZAYDIN Spring Term-2017 Lecture 2 Analog Sound to Digital Sound Characteristics of Sound Amplitude Wavelength (w) Frequency ( ) Timbre
More informationEnhanced Waveform Interpolative Coding at 4 kbps
Enhanced Waveform Interpolative Coding at 4 kbps Oded Gottesman, and Allen Gersho Signal Compression Lab. University of California, Santa Barbara E-mail: [oded, gersho]@scl.ece.ucsb.edu Signal Compression
More informationLecture Schedule: Week Date Lecture Title
http://elec3004.org Sampling & More 2014 School of Information Technology and Electrical Engineering at The University of Queensland Lecture Schedule: Week Date Lecture Title 1 2-Mar Introduction 3-Mar
More informationMachine recognition of speech trained on data from New Jersey Labs
Machine recognition of speech trained on data from New Jersey Labs Frequency response (peak around 5 Hz) Impulse response (effective length around 200 ms) 41 RASTA filter 10 attenuation [db] 40 1 10 modulation
More informationSpeech Coding using Linear Prediction
Speech Coding using Linear Prediction Jesper Kjær Nielsen Aalborg University and Bang & Olufsen jkn@es.aau.dk September 10, 2015 1 Background Speech is generated when air is pushed from the lungs through
More informationADDITIVE SYNTHESIS BASED ON THE CONTINUOUS WAVELET TRANSFORM: A SINUSOIDAL PLUS TRANSIENT MODEL
ADDITIVE SYNTHESIS BASED ON THE CONTINUOUS WAVELET TRANSFORM: A SINUSOIDAL PLUS TRANSIENT MODEL José R. Beltrán and Fernando Beltrán Department of Electronic Engineering and Communications University of
More informationIsolated Word Recognition Based on Combination of Multiple Noise-Robust Techniques
Isolated Word Recognition Based on Combination of Multiple Noise-Robust Techniques 81 Isolated Word Recognition Based on Combination of Multiple Noise-Robust Techniques Noboru Hayasaka 1, Non-member ABSTRACT
More informationRobust telephone speech recognition based on channel compensation
Pattern Recognition 32 (1999) 1061}1067 Robust telephone speech recognition based on channel compensation Jiqing Han*, Wen Gao Department of Computer Science and Engineering, Harbin Institute of Technology,
More informationEffective post-processing for single-channel frequency-domain speech enhancement Weifeng Li a
R E S E A R C H R E P O R T I D I A P Effective post-processing for single-channel frequency-domain speech enhancement Weifeng Li a IDIAP RR 7-7 January 8 submitted for publication a IDIAP Research Institute,
More informationSGN Audio and Speech Processing
Introduction 1 Course goals Introduction 2 SGN 14006 Audio and Speech Processing Lectures, Fall 2014 Anssi Klapuri Tampere University of Technology! Learn basics of audio signal processing Basic operations
More informationEE216B: VLSI Signal Processing. Wavelets. Prof. Dejan Marković Shortcomings of the Fourier Transform (FT)
5//0 EE6B: VLSI Signal Processing Wavelets Prof. Dejan Marković ee6b@gmail.com Shortcomings of the Fourier Transform (FT) FT gives information about the spectral content of the signal but loses all time
More informationEffects of Reverberation on Pitch, Onset/Offset, and Binaural Cues
Effects of Reverberation on Pitch, Onset/Offset, and Binaural Cues DeLiang Wang Perception & Neurodynamics Lab The Ohio State University Outline of presentation Introduction Human performance Reverberation
More informationSignal Analysis Using Autoregressive Models of Amplitude Modulation. Sriram Ganapathy
Signal Analysis Using Autoregressive Models of Amplitude Modulation Sriram Ganapathy Advisor - Hynek Hermansky Johns Hopkins University 11-18-2011 Overview Introduction AR Model of Hilbert Envelopes FDLP
More informationAudio Signal Compression using DCT and LPC Techniques
Audio Signal Compression using DCT and LPC Techniques P. Sandhya Rani#1, D.Nanaji#2, V.Ramesh#3,K.V.S. Kiran#4 #Student, Department of ECE, Lendi Institute Of Engineering And Technology, Vizianagaram,
More informationReading: Johnson Ch , Ch.5.5 (today); Liljencrants & Lindblom; Stevens (Tues) reminder: no class on Thursday.
L105/205 Phonetics Scarborough Handout 7 10/18/05 Reading: Johnson Ch.2.3.3-2.3.6, Ch.5.5 (today); Liljencrants & Lindblom; Stevens (Tues) reminder: no class on Thursday Spectral Analysis 1. There are
More informationBiomedical Signals. Signals and Images in Medicine Dr Nabeel Anwar
Biomedical Signals Signals and Images in Medicine Dr Nabeel Anwar Noise Removal: Time Domain Techniques 1. Synchronized Averaging (covered in lecture 1) 2. Moving Average Filters (today s topic) 3. Derivative
More informationSignals and Systems Lecture 6: Fourier Applications
Signals and Systems Lecture 6: Fourier Applications Farzaneh Abdollahi Department of Electrical Engineering Amirkabir University of Technology Winter 2012 arzaneh Abdollahi Signal and Systems Lecture 6
More informationRelative phase information for detecting human speech and spoofed speech
Relative phase information for detecting human speech and spoofed speech Longbiao Wang 1, Yohei Yoshida 1, Yuta Kawakami 1 and Seiichi Nakagawa 2 1 Nagaoka University of Technology, Japan 2 Toyohashi University
More informationSignals and Systems Lecture 6: Fourier Applications
Signals and Systems Lecture 6: Fourier Applications Farzaneh Abdollahi Department of Electrical Engineering Amirkabir University of Technology Winter 2012 arzaneh Abdollahi Signal and Systems Lecture 6
More informationSound Synthesis Methods
Sound Synthesis Methods Matti Vihola, mvihola@cs.tut.fi 23rd August 2001 1 Objectives The objective of sound synthesis is to create sounds that are Musically interesting Preferably realistic (sounds like
More informationApplications of Music Processing
Lecture Music Processing Applications of Music Processing Christian Dittmar International Audio Laboratories Erlangen christian.dittmar@audiolabs-erlangen.de Singing Voice Detection Important pre-requisite
More informationEnhancement of Speech Signal by Adaptation of Scales and Thresholds of Bionic Wavelet Transform Coefficients
ISSN (Print) : 232 3765 An ISO 3297: 27 Certified Organization Vol. 3, Special Issue 3, April 214 Paiyanoor-63 14, Tamil Nadu, India Enhancement of Speech Signal by Adaptation of Scales and Thresholds
More informationPattern Recognition. Part 6: Bandwidth Extension. Gerhard Schmidt
Pattern Recognition Part 6: Gerhard Schmidt Christian-Albrechts-Universität zu Kiel Faculty of Engineering Institute of Electrical and Information Engineering Digital Signal Processing and System Theory
More informationSpeech Compression Using Voice Excited Linear Predictive Coding
Speech Compression Using Voice Excited Linear Predictive Coding Ms.Tosha Sen, Ms.Kruti Jay Pancholi PG Student, Asst. Professor, L J I E T, Ahmedabad Abstract : The aim of the thesis is design good quality
More informationSpeech Enhancement Based On Spectral Subtraction For Speech Recognition System With Dpcm
International OPEN ACCESS Journal Of Modern Engineering Research (IJMER) Speech Enhancement Based On Spectral Subtraction For Speech Recognition System With Dpcm A.T. Rajamanickam, N.P.Subiramaniyam, A.Balamurugan*,
More informationLearning to Unlearn and Relearn Speech Signal Processing using Neural Networks: current and future perspectives
Learning to Unlearn and Relearn Speech Signal Processing using Neural Networks: current and future perspectives Mathew Magimai Doss Collaborators: Vinayak Abrol, Selen Hande Kabil, Hannah Muckenhirn, Dimitri
More information