SlideShare a Scribd company logo
1 of 19
Download to read offline
A Geometrically Constrained Independent Vector Analysis
Algorithm for Online Source Extraction
Affan Khan, Maja Taseska, Emanu¨el Habets
affan.khan@audiolabs-erlangen.de
September 7, 2015
Content
1. Introduction
Motivation
State of the art vs Proposed algorithm
Signal model
Independent Vector Analysis (IVA)
2. Constrained IVA (CIVA)
3. Results
4. Conclusions
© AudioLabs 2015
Slide 1
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Introduction
Motivation
Many applications require a handsfree capture of speech using one or more
microphones.
Signal received at the microphones is usually a mixture of desired and
undesired source signals.
Extraction of desired source from the mixture is required.
© AudioLabs 2015
Slide 2
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Introduction
State of the art
Beamforming algorithms
Commonly used to extract one source signal.
Beamformers can either be data-independent or data-dependent.
Data-dependent beamformers require an accurate estimate of SOS.
Independent component analysis (ICA)
Goal is to extract all source signals.
Assumption of statistical independence.
Limited performance in reverberant and non-determined scenarios.
Researchers incorporated prior knowledge in ICA to improve performance.
© AudioLabs 2015
Slide 3
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Introduction
State of the art vs Proposed algorithm
Geometric prior information can be added to ICA through soft and hard
constrained optimization [1-2].
These algorithms require prior knowledge on the number of sources.
The algorithms assume all samples are available (batch mode).
Permutation ambiguity not completely resolved.
Proposed algorithm
No prior knowledge on the number of sources required.
An online algorithm.
No permutation ambiguity due to the use of IVA instead of ICA.
[1] Parra and Alvino ”Geometric Source Separation: Merging Convolutive Source Separation With Geometric
Beamforming”, IEEE Transactions on Speech and Audio Processing, 2002, pp. 352-362.
[2] Knaak, Araki and Makino ”Geometrically Constrained Independent Component Analysis”, IEEE Transactions on Audio,
Speech and Language Processing, 2007, pp. 715-726.
© AudioLabs 2015
Slide 4
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Introduction
Signal model
The mixture signal received at the m − th microphone
Ym(n, k) =
L
l=1
Am,l(k) Sl(n, k) + Vm(n, k)
We consider the problem where only one of the L sources is desired
y(n, k) = a1(k) S1(n, k) +
L
u=2
au(k) Su(n, k) + v(n, k)
Extraction of source signals from the received mixture in a standard BSS
algorithm is achieved by a demixing matrix W(k)
ˆs(n, k) = W(k) y(n, k)
© AudioLabs 2015
Slide 5
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Introduction
Signal model
The demixing matrix W(k) can be written out as follows
W(k) = [w1(k) w2(k) w3(k) · · · wM (k)]H
Goal: Given the DOA of the desired source, compute a demixing matrix
W(k) while ensuring that w1(k) extracts the signal originating from the
desired direction.
© AudioLabs 2015
Slide 6
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Introduction
Independent Vector Analysis (IVA)
Computes the demixing matrix through minimization of statistical
dependence between the output signals across all frequency bins.
Jiva = KL p ˆS1(n) · · · ˆSM (n) ||
M
m=1
p ˆSm(n)
Jiva = −
M
m=1
E log p ˆSm(n) −
K
k=1
log|det [W(k)] |
An online variant of IVA was proposed in [3].
[3] Kim ”Real-Time Independent Vector Analysis for Convolutive Blind Source Separation”, IEEE Transactions on Circuits
and Systems 1, 2010, pp. 1431-1438.
© AudioLabs 2015
Slide 7
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Content
1. Introduction
2. Constrained IVA (CIVA)
IVA with geometric constraints
Online CIVA algorithm
3. Results
4. Conclusions
© AudioLabs 2015
Slide 8
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Constrained IVA (CIVA)
IVA with geometric constraints
Given the DOA of the desired source, we constrain the look direction of
w1(k) using RTF vector under far field assumption
g1(k) = 1 ej(2πf/c)[r2 − r1]T
q1
· · · ej(2πf/c)[rM − r1]T
q1
T
where q1 is a unit vector pointing in the direction of the desired source.
A penalty function to restrict the Euclidean angle between w1(k) and g1(k)
Jp =
K
k=1
[cos Θ(k) − 1]2
cos Θ(k) =
Re wH
1 (k) g1(k)
||w1(k)|| ||g1(k)||
© AudioLabs 2015
Slide 9
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Constrained IVA (CIVA)
IVA with geometric constraints
The cost function of IVA with geometric constraints can be expressed as
Jciva = Jiva + λ Jp
The gradient of Jciva with respect to the elements of the demixing matrix
Wciva(k) = E{ϕ(k)
(n)y
H
(n, k)} − W
−H
(k)
Wiva(k)
+λ Wp(k)
where
Wp(k) = C · (cos Θ(k) − 1) g1(k) −
w1(k)
||w1(k)||2
Re w
H
1 (k)g1(k)
© AudioLabs 2015
Slide 10
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Constrained IVA (CIVA)
An online CIVA algorithm
An online variant of the CIVA algorithm is obtained by omitting the
expectation operator in the calculation of Wiva(k).
To avoid divergence of the algorithm due to source signal fluctuations, we
normalize the gradient matrix at each frame by its Frobenius norm || · ||F
Wn(k) = Wn−1(k) − η
Wn,civa(k)
|| Wn,civa(k)||F
© AudioLabs 2015
Slide 11
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Content
1. Introduction
2. Constrained IVA (CIVA)
3. Results
4. Conclusions
© AudioLabs 2015
Slide 12
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Results: Constrained IVA
Experimental setup
Dimensions of the simulated room: 7.5m × 5.5m × 3m.
Uniform circular array with a diameter of 2.5 cm.
Sampling rate 16 kHz, STFT frame 1024 samples (50% overlap).
Diffuse noise 30dB SDR, uncorrelated sensor noise 50dB SNR.
Segmental speech distortion index (SegSD) and segmental signal to
interference ratio (SegSIR) are used as performance measures.
© AudioLabs 2015
Slide 13
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Results
Fixed number of active sources
Proposed constrained IVA algorithm converges in all scenarios.
Provides low desired speech signal distortion.
T60 = 150 ms T60 = 300 ms
segSD segSIR (dB) segSD segSIR (dB)
Unprocessed mixture 1.9 1.5
IVA (M = 3, L = 3) 0.21 14.9 0.31 4.7
CIVA (M = 3, L = 3) 0.12 11.7 0.24 5.2
Unprocessed mixture 1.9 1.5
IVA (M = 4, L = 3) 0.27 8.2 0.39 3.8
CIVA (M = 4, L = 3) 0.04 17.0 0.07 8.7
Unprocessed mixture 0.7 -0.3
IVA (M = 4, L = 5) 0.54 0.2 0.57 0.1
CIVA (M = 4, L = 5) 0.05 8.7 0.08 3.6
Table: Performance measures for fixed number of sources
© AudioLabs 2015
Slide 14
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Results
Time-varying number of active sources
Figure: Top: Activity pattern of speech sources over time; Bottom: Segmental SIR
improvement over time. M = 6, T60 = 150 ms.
© AudioLabs 2015
Slide 15
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Content
1. Introduction
2. Constrained IVA (CIVA)
3. Results
4. Conclusions
© AudioLabs 2015
Slide 16
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Conclusions
An geometrically constrained online IVA algorithm was proposed to extract
the desired speech signal.
The applicability of the algorithm to scenarios with fixed and time varying
number of interferers was demonstrated.
Future work includes comparison of the algorithm against beamforming
algorithms and evaluation with recorded data.
© AudioLabs 2015
Slide 17
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets
Conclusions
An geometrically constrained online IVA algorithm was proposed to extract
the desired speech signal.
The applicability of the algorithm to scenarios with fixed and time varying
number of interferers was demonstrated.
Future work includes comparison of the algorithm against beamforming
algorithms and evaluation with recorded data.
Thank you for your attention!
© AudioLabs 2015
Slide 17
Constrained Independent Vector Analysis
Affan Khan, Maja Taseska, Emanu¨el Habets

More Related Content

What's hot

エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 2)~
エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 2)~エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 2)~
エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 2)~Yamagishi Laboratory, National Institute of Informatics, Japan
 
Wave File Features Extraction using Reduced LBP
Wave File Features Extraction using Reduced LBP Wave File Features Extraction using Reduced LBP
Wave File Features Extraction using Reduced LBP IJECEIAES
 
An Efficient DSP Implementation of a Dynamic Convolution Using Principal Comp...
An Efficient DSP Implementation of a Dynamic Convolution Using Principal Comp...An Efficient DSP Implementation of a Dynamic Convolution Using Principal Comp...
An Efficient DSP Implementation of a Dynamic Convolution Using Principal Comp...a3labdsp
 
Subjective comparison of_speech_enhancement_algori (1)
Subjective comparison of_speech_enhancement_algori (1)Subjective comparison of_speech_enhancement_algori (1)
Subjective comparison of_speech_enhancement_algori (1)Priyanka Reddy
 
PERFORMANCE ANALYIS OF LMS ADAPTIVE FIR FILTER AND RLS ADAPTIVE FIR FILTER FO...
PERFORMANCE ANALYIS OF LMS ADAPTIVE FIR FILTER AND RLS ADAPTIVE FIR FILTER FO...PERFORMANCE ANALYIS OF LMS ADAPTIVE FIR FILTER AND RLS ADAPTIVE FIR FILTER FO...
PERFORMANCE ANALYIS OF LMS ADAPTIVE FIR FILTER AND RLS ADAPTIVE FIR FILTER FO...sipij
 
ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]Dongmin Choi
 
SERENE 2014 School: Daniel varro serene2014_school
SERENE 2014 School: Daniel varro serene2014_schoolSERENE 2014 School: Daniel varro serene2014_school
SERENE 2014 School: Daniel varro serene2014_schoolHenry Muccini
 
An Efficient DSP Based Implementation of a Fast Convolution Approach with non...
An Efficient DSP Based Implementation of a Fast Convolution Approach with non...An Efficient DSP Based Implementation of a Fast Convolution Approach with non...
An Efficient DSP Based Implementation of a Fast Convolution Approach with non...a3labdsp
 
Introduction to deep learning based voice activity detection
Introduction to deep learning based voice activity detectionIntroduction to deep learning based voice activity detection
Introduction to deep learning based voice activity detectionNAVER Engineering
 
FPGA IMPLEMENTATION OF NOISE CANCELLATION USING ADAPTIVE ALGORITHMS
FPGA IMPLEMENTATION OF NOISE CANCELLATION USING ADAPTIVE ALGORITHMSFPGA IMPLEMENTATION OF NOISE CANCELLATION USING ADAPTIVE ALGORITHMS
FPGA IMPLEMENTATION OF NOISE CANCELLATION USING ADAPTIVE ALGORITHMSEditor IJMTER
 
Deformable DETR Review [CDM]
Deformable DETR Review [CDM]Deformable DETR Review [CDM]
Deformable DETR Review [CDM]Dongmin Choi
 
Real-Time Active Noise Cancellation with Simulink and Data Acquisition Toolbox
Real-Time Active Noise Cancellation with Simulink and Data Acquisition ToolboxReal-Time Active Noise Cancellation with Simulink and Data Acquisition Toolbox
Real-Time Active Noise Cancellation with Simulink and Data Acquisition ToolboxIDES Editor
 
IFIP Networking 2015
IFIP Networking 2015IFIP Networking 2015
IFIP Networking 2015SmartenIT
 
Real time active noise cancellation using adaptive filters following RLS and ...
Real time active noise cancellation using adaptive filters following RLS and ...Real time active noise cancellation using adaptive filters following RLS and ...
Real time active noise cancellation using adaptive filters following RLS and ...IRJET Journal
 
Hardware Implementation of Adaptive Noise Cancellation over DSP Kit TMS320C6713
Hardware Implementation of Adaptive Noise Cancellation over DSP Kit TMS320C6713Hardware Implementation of Adaptive Noise Cancellation over DSP Kit TMS320C6713
Hardware Implementation of Adaptive Noise Cancellation over DSP Kit TMS320C6713CSCJournals
 

What's hot (20)

エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 2)~
エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 2)~エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 2)~
エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 2)~
 
Wave File Features Extraction using Reduced LBP
Wave File Features Extraction using Reduced LBP Wave File Features Extraction using Reduced LBP
Wave File Features Extraction using Reduced LBP
 
An Efficient DSP Implementation of a Dynamic Convolution Using Principal Comp...
An Efficient DSP Implementation of a Dynamic Convolution Using Principal Comp...An Efficient DSP Implementation of a Dynamic Convolution Using Principal Comp...
An Efficient DSP Implementation of a Dynamic Convolution Using Principal Comp...
 
Neural Waveform Modeling
Neural Waveform Modeling Neural Waveform Modeling
Neural Waveform Modeling
 
Subjective comparison of_speech_enhancement_algori (1)
Subjective comparison of_speech_enhancement_algori (1)Subjective comparison of_speech_enhancement_algori (1)
Subjective comparison of_speech_enhancement_algori (1)
 
PERFORMANCE ANALYIS OF LMS ADAPTIVE FIR FILTER AND RLS ADAPTIVE FIR FILTER FO...
PERFORMANCE ANALYIS OF LMS ADAPTIVE FIR FILTER AND RLS ADAPTIVE FIR FILTER FO...PERFORMANCE ANALYIS OF LMS ADAPTIVE FIR FILTER AND RLS ADAPTIVE FIR FILTER FO...
PERFORMANCE ANALYIS OF LMS ADAPTIVE FIR FILTER AND RLS ADAPTIVE FIR FILTER FO...
 
ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]
 
Visual Object Analysis using Regions and Local Features
Visual Object Analysis using Regions and Local FeaturesVisual Object Analysis using Regions and Local Features
Visual Object Analysis using Regions and Local Features
 
SERENE 2014 School: Daniel varro serene2014_school
SERENE 2014 School: Daniel varro serene2014_schoolSERENE 2014 School: Daniel varro serene2014_school
SERENE 2014 School: Daniel varro serene2014_school
 
An Efficient DSP Based Implementation of a Fast Convolution Approach with non...
An Efficient DSP Based Implementation of a Fast Convolution Approach with non...An Efficient DSP Based Implementation of a Fast Convolution Approach with non...
An Efficient DSP Based Implementation of a Fast Convolution Approach with non...
 
Introduction to deep learning based voice activity detection
Introduction to deep learning based voice activity detectionIntroduction to deep learning based voice activity detection
Introduction to deep learning based voice activity detection
 
FPGA IMPLEMENTATION OF NOISE CANCELLATION USING ADAPTIVE ALGORITHMS
FPGA IMPLEMENTATION OF NOISE CANCELLATION USING ADAPTIVE ALGORITHMSFPGA IMPLEMENTATION OF NOISE CANCELLATION USING ADAPTIVE ALGORITHMS
FPGA IMPLEMENTATION OF NOISE CANCELLATION USING ADAPTIVE ALGORITHMS
 
Tutorial on end-to-end text-to-speech synthesis: Part 2 – Tactron and related...
Tutorial on end-to-end text-to-speech synthesis: Part 2 – Tactron and related...Tutorial on end-to-end text-to-speech synthesis: Part 2 – Tactron and related...
Tutorial on end-to-end text-to-speech synthesis: Part 2 – Tactron and related...
 
ANCLMS
ANCLMSANCLMS
ANCLMS
 
Deformable DETR Review [CDM]
Deformable DETR Review [CDM]Deformable DETR Review [CDM]
Deformable DETR Review [CDM]
 
Real-Time Active Noise Cancellation with Simulink and Data Acquisition Toolbox
Real-Time Active Noise Cancellation with Simulink and Data Acquisition ToolboxReal-Time Active Noise Cancellation with Simulink and Data Acquisition Toolbox
Real-Time Active Noise Cancellation with Simulink and Data Acquisition Toolbox
 
Adaptive filter
Adaptive filterAdaptive filter
Adaptive filter
 
IFIP Networking 2015
IFIP Networking 2015IFIP Networking 2015
IFIP Networking 2015
 
Real time active noise cancellation using adaptive filters following RLS and ...
Real time active noise cancellation using adaptive filters following RLS and ...Real time active noise cancellation using adaptive filters following RLS and ...
Real time active noise cancellation using adaptive filters following RLS and ...
 
Hardware Implementation of Adaptive Noise Cancellation over DSP Kit TMS320C6713
Hardware Implementation of Adaptive Noise Cancellation over DSP Kit TMS320C6713Hardware Implementation of Adaptive Noise Cancellation over DSP Kit TMS320C6713
Hardware Implementation of Adaptive Noise Cancellation over DSP Kit TMS320C6713
 

Similar to Geometrically Constrained Independent Vector Analysis

Text Independent Speaker Identification Using Imfcc Integrated With Ica
Text Independent Speaker Identification Using Imfcc Integrated With IcaText Independent Speaker Identification Using Imfcc Integrated With Ica
Text Independent Speaker Identification Using Imfcc Integrated With IcaIOSR Journals
 
Single Channel Speech De-noising Using Kernel Independent Component Analysis...
	Single Channel Speech De-noising Using Kernel Independent Component Analysis...	Single Channel Speech De-noising Using Kernel Independent Component Analysis...
Single Channel Speech De-noising Using Kernel Independent Component Analysis...theijes
 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...ijceronline
 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...ijceronline
 
Adaptive Channel Equalization using Multilayer Perceptron Neural Networks wit...
Adaptive Channel Equalization using Multilayer Perceptron Neural Networks wit...Adaptive Channel Equalization using Multilayer Perceptron Neural Networks wit...
Adaptive Channel Equalization using Multilayer Perceptron Neural Networks wit...IOSRJVSP
 
Analysis the results_of_acoustic_echo_cancellation_for_speech_processing_usin...
Analysis the results_of_acoustic_echo_cancellation_for_speech_processing_usin...Analysis the results_of_acoustic_echo_cancellation_for_speech_processing_usin...
Analysis the results_of_acoustic_echo_cancellation_for_speech_processing_usin...Venkata Sudhir Vedurla
 
A Novel CAZAC Sequence Based Timing Synchronization Scheme for OFDM System
A Novel CAZAC Sequence Based Timing Synchronization Scheme for OFDM SystemA Novel CAZAC Sequence Based Timing Synchronization Scheme for OFDM System
A Novel CAZAC Sequence Based Timing Synchronization Scheme for OFDM SystemIJAAS Team
 
Performance Evaluation of Conventional and Hybrid Feature Extractions Using M...
Performance Evaluation of Conventional and Hybrid Feature Extractions Using M...Performance Evaluation of Conventional and Hybrid Feature Extractions Using M...
Performance Evaluation of Conventional and Hybrid Feature Extractions Using M...IJERA Editor
 
A Noise Reduction Method Based on Modified Least Mean Square Algorithm of Rea...
A Noise Reduction Method Based on Modified Least Mean Square Algorithm of Rea...A Noise Reduction Method Based on Modified Least Mean Square Algorithm of Rea...
A Noise Reduction Method Based on Modified Least Mean Square Algorithm of Rea...IRJET Journal
 
A New Method for a Nonlinear Acoustic Echo Cancellation System
A New Method for a Nonlinear Acoustic Echo Cancellation SystemA New Method for a Nonlinear Acoustic Echo Cancellation System
A New Method for a Nonlinear Acoustic Echo Cancellation SystemIRJET Journal
 
Performance Evaluation of Adaptive Filters Structures for Acoustic Echo Cance...
Performance Evaluation of Adaptive Filters Structures for Acoustic Echo Cance...Performance Evaluation of Adaptive Filters Structures for Acoustic Echo Cance...
Performance Evaluation of Adaptive Filters Structures for Acoustic Echo Cance...CSCJournals
 
IRJET- Artificial Neural Network Algorithm for Acoustic Echo Cancellation App...
IRJET- Artificial Neural Network Algorithm for Acoustic Echo Cancellation App...IRJET- Artificial Neural Network Algorithm for Acoustic Echo Cancellation App...
IRJET- Artificial Neural Network Algorithm for Acoustic Echo Cancellation App...IRJET Journal
 
IRJET- A Novel Adaptive Sub-Band Filter Design with BD-VSS using Particle Swa...
IRJET- A Novel Adaptive Sub-Band Filter Design with BD-VSS using Particle Swa...IRJET- A Novel Adaptive Sub-Band Filter Design with BD-VSS using Particle Swa...
IRJET- A Novel Adaptive Sub-Band Filter Design with BD-VSS using Particle Swa...IRJET Journal
 
performance analysis of MUSIC and ESPRIT DOA estimation used in adaptive arra...
performance analysis of MUSIC and ESPRIT DOA estimation used in adaptive arra...performance analysis of MUSIC and ESPRIT DOA estimation used in adaptive arra...
performance analysis of MUSIC and ESPRIT DOA estimation used in adaptive arra...CSCJournals
 
Adaptive Channel Equalization for Nonlinear Channels using Signed Regressor F...
Adaptive Channel Equalization for Nonlinear Channels using Signed Regressor F...Adaptive Channel Equalization for Nonlinear Channels using Signed Regressor F...
Adaptive Channel Equalization for Nonlinear Channels using Signed Regressor F...IDES Editor
 
Implementation of Interleaving Methods on MELP 2.4 Coder to Reduce Packet Los...
Implementation of Interleaving Methods on MELP 2.4 Coder to Reduce Packet Los...Implementation of Interleaving Methods on MELP 2.4 Coder to Reduce Packet Los...
Implementation of Interleaving Methods on MELP 2.4 Coder to Reduce Packet Los...IJERA Editor
 
Adaptive Noise Cancellation using Multirate Techniques
Adaptive Noise Cancellation using Multirate TechniquesAdaptive Noise Cancellation using Multirate Techniques
Adaptive Noise Cancellation using Multirate TechniquesIJERD Editor
 
IRJET- Compressed Sensing based Modified Orthogonal Matching Pursuit in DTTV ...
IRJET- Compressed Sensing based Modified Orthogonal Matching Pursuit in DTTV ...IRJET- Compressed Sensing based Modified Orthogonal Matching Pursuit in DTTV ...
IRJET- Compressed Sensing based Modified Orthogonal Matching Pursuit in DTTV ...IRJET Journal
 
Direction of Arrival Estimation Based on MUSIC Algorithm Using Uniform and No...
Direction of Arrival Estimation Based on MUSIC Algorithm Using Uniform and No...Direction of Arrival Estimation Based on MUSIC Algorithm Using Uniform and No...
Direction of Arrival Estimation Based on MUSIC Algorithm Using Uniform and No...IJERA Editor
 

Similar to Geometrically Constrained Independent Vector Analysis (20)

Text Independent Speaker Identification Using Imfcc Integrated With Ica
Text Independent Speaker Identification Using Imfcc Integrated With IcaText Independent Speaker Identification Using Imfcc Integrated With Ica
Text Independent Speaker Identification Using Imfcc Integrated With Ica
 
Single Channel Speech De-noising Using Kernel Independent Component Analysis...
	Single Channel Speech De-noising Using Kernel Independent Component Analysis...	Single Channel Speech De-noising Using Kernel Independent Component Analysis...
Single Channel Speech De-noising Using Kernel Independent Component Analysis...
 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
 
Adaptive Channel Equalization using Multilayer Perceptron Neural Networks wit...
Adaptive Channel Equalization using Multilayer Perceptron Neural Networks wit...Adaptive Channel Equalization using Multilayer Perceptron Neural Networks wit...
Adaptive Channel Equalization using Multilayer Perceptron Neural Networks wit...
 
Analysis the results_of_acoustic_echo_cancellation_for_speech_processing_usin...
Analysis the results_of_acoustic_echo_cancellation_for_speech_processing_usin...Analysis the results_of_acoustic_echo_cancellation_for_speech_processing_usin...
Analysis the results_of_acoustic_echo_cancellation_for_speech_processing_usin...
 
A Novel CAZAC Sequence Based Timing Synchronization Scheme for OFDM System
A Novel CAZAC Sequence Based Timing Synchronization Scheme for OFDM SystemA Novel CAZAC Sequence Based Timing Synchronization Scheme for OFDM System
A Novel CAZAC Sequence Based Timing Synchronization Scheme for OFDM System
 
Performance Evaluation of Conventional and Hybrid Feature Extractions Using M...
Performance Evaluation of Conventional and Hybrid Feature Extractions Using M...Performance Evaluation of Conventional and Hybrid Feature Extractions Using M...
Performance Evaluation of Conventional and Hybrid Feature Extractions Using M...
 
A Noise Reduction Method Based on Modified Least Mean Square Algorithm of Rea...
A Noise Reduction Method Based on Modified Least Mean Square Algorithm of Rea...A Noise Reduction Method Based on Modified Least Mean Square Algorithm of Rea...
A Noise Reduction Method Based on Modified Least Mean Square Algorithm of Rea...
 
A New Method for a Nonlinear Acoustic Echo Cancellation System
A New Method for a Nonlinear Acoustic Echo Cancellation SystemA New Method for a Nonlinear Acoustic Echo Cancellation System
A New Method for a Nonlinear Acoustic Echo Cancellation System
 
Performance Evaluation of Adaptive Filters Structures for Acoustic Echo Cance...
Performance Evaluation of Adaptive Filters Structures for Acoustic Echo Cance...Performance Evaluation of Adaptive Filters Structures for Acoustic Echo Cance...
Performance Evaluation of Adaptive Filters Structures for Acoustic Echo Cance...
 
IRJET- Artificial Neural Network Algorithm for Acoustic Echo Cancellation App...
IRJET- Artificial Neural Network Algorithm for Acoustic Echo Cancellation App...IRJET- Artificial Neural Network Algorithm for Acoustic Echo Cancellation App...
IRJET- Artificial Neural Network Algorithm for Acoustic Echo Cancellation App...
 
IRJET- A Novel Adaptive Sub-Band Filter Design with BD-VSS using Particle Swa...
IRJET- A Novel Adaptive Sub-Band Filter Design with BD-VSS using Particle Swa...IRJET- A Novel Adaptive Sub-Band Filter Design with BD-VSS using Particle Swa...
IRJET- A Novel Adaptive Sub-Band Filter Design with BD-VSS using Particle Swa...
 
performance analysis of MUSIC and ESPRIT DOA estimation used in adaptive arra...
performance analysis of MUSIC and ESPRIT DOA estimation used in adaptive arra...performance analysis of MUSIC and ESPRIT DOA estimation used in adaptive arra...
performance analysis of MUSIC and ESPRIT DOA estimation used in adaptive arra...
 
Adaptive Channel Equalization for Nonlinear Channels using Signed Regressor F...
Adaptive Channel Equalization for Nonlinear Channels using Signed Regressor F...Adaptive Channel Equalization for Nonlinear Channels using Signed Regressor F...
Adaptive Channel Equalization for Nonlinear Channels using Signed Regressor F...
 
F43063841
F43063841F43063841
F43063841
 
Implementation of Interleaving Methods on MELP 2.4 Coder to Reduce Packet Los...
Implementation of Interleaving Methods on MELP 2.4 Coder to Reduce Packet Los...Implementation of Interleaving Methods on MELP 2.4 Coder to Reduce Packet Los...
Implementation of Interleaving Methods on MELP 2.4 Coder to Reduce Packet Los...
 
Adaptive Noise Cancellation using Multirate Techniques
Adaptive Noise Cancellation using Multirate TechniquesAdaptive Noise Cancellation using Multirate Techniques
Adaptive Noise Cancellation using Multirate Techniques
 
IRJET- Compressed Sensing based Modified Orthogonal Matching Pursuit in DTTV ...
IRJET- Compressed Sensing based Modified Orthogonal Matching Pursuit in DTTV ...IRJET- Compressed Sensing based Modified Orthogonal Matching Pursuit in DTTV ...
IRJET- Compressed Sensing based Modified Orthogonal Matching Pursuit in DTTV ...
 
Direction of Arrival Estimation Based on MUSIC Algorithm Using Uniform and No...
Direction of Arrival Estimation Based on MUSIC Algorithm Using Uniform and No...Direction of Arrival Estimation Based on MUSIC Algorithm Using Uniform and No...
Direction of Arrival Estimation Based on MUSIC Algorithm Using Uniform and No...
 

Geometrically Constrained Independent Vector Analysis

  • 1. A Geometrically Constrained Independent Vector Analysis Algorithm for Online Source Extraction Affan Khan, Maja Taseska, Emanu¨el Habets affan.khan@audiolabs-erlangen.de September 7, 2015
  • 2. Content 1. Introduction Motivation State of the art vs Proposed algorithm Signal model Independent Vector Analysis (IVA) 2. Constrained IVA (CIVA) 3. Results 4. Conclusions © AudioLabs 2015 Slide 1 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 3. Introduction Motivation Many applications require a handsfree capture of speech using one or more microphones. Signal received at the microphones is usually a mixture of desired and undesired source signals. Extraction of desired source from the mixture is required. © AudioLabs 2015 Slide 2 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 4. Introduction State of the art Beamforming algorithms Commonly used to extract one source signal. Beamformers can either be data-independent or data-dependent. Data-dependent beamformers require an accurate estimate of SOS. Independent component analysis (ICA) Goal is to extract all source signals. Assumption of statistical independence. Limited performance in reverberant and non-determined scenarios. Researchers incorporated prior knowledge in ICA to improve performance. © AudioLabs 2015 Slide 3 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 5. Introduction State of the art vs Proposed algorithm Geometric prior information can be added to ICA through soft and hard constrained optimization [1-2]. These algorithms require prior knowledge on the number of sources. The algorithms assume all samples are available (batch mode). Permutation ambiguity not completely resolved. Proposed algorithm No prior knowledge on the number of sources required. An online algorithm. No permutation ambiguity due to the use of IVA instead of ICA. [1] Parra and Alvino ”Geometric Source Separation: Merging Convolutive Source Separation With Geometric Beamforming”, IEEE Transactions on Speech and Audio Processing, 2002, pp. 352-362. [2] Knaak, Araki and Makino ”Geometrically Constrained Independent Component Analysis”, IEEE Transactions on Audio, Speech and Language Processing, 2007, pp. 715-726. © AudioLabs 2015 Slide 4 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 6. Introduction Signal model The mixture signal received at the m − th microphone Ym(n, k) = L l=1 Am,l(k) Sl(n, k) + Vm(n, k) We consider the problem where only one of the L sources is desired y(n, k) = a1(k) S1(n, k) + L u=2 au(k) Su(n, k) + v(n, k) Extraction of source signals from the received mixture in a standard BSS algorithm is achieved by a demixing matrix W(k) ˆs(n, k) = W(k) y(n, k) © AudioLabs 2015 Slide 5 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 7. Introduction Signal model The demixing matrix W(k) can be written out as follows W(k) = [w1(k) w2(k) w3(k) · · · wM (k)]H Goal: Given the DOA of the desired source, compute a demixing matrix W(k) while ensuring that w1(k) extracts the signal originating from the desired direction. © AudioLabs 2015 Slide 6 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 8. Introduction Independent Vector Analysis (IVA) Computes the demixing matrix through minimization of statistical dependence between the output signals across all frequency bins. Jiva = KL p ˆS1(n) · · · ˆSM (n) || M m=1 p ˆSm(n) Jiva = − M m=1 E log p ˆSm(n) − K k=1 log|det [W(k)] | An online variant of IVA was proposed in [3]. [3] Kim ”Real-Time Independent Vector Analysis for Convolutive Blind Source Separation”, IEEE Transactions on Circuits and Systems 1, 2010, pp. 1431-1438. © AudioLabs 2015 Slide 7 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 9. Content 1. Introduction 2. Constrained IVA (CIVA) IVA with geometric constraints Online CIVA algorithm 3. Results 4. Conclusions © AudioLabs 2015 Slide 8 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 10. Constrained IVA (CIVA) IVA with geometric constraints Given the DOA of the desired source, we constrain the look direction of w1(k) using RTF vector under far field assumption g1(k) = 1 ej(2πf/c)[r2 − r1]T q1 · · · ej(2πf/c)[rM − r1]T q1 T where q1 is a unit vector pointing in the direction of the desired source. A penalty function to restrict the Euclidean angle between w1(k) and g1(k) Jp = K k=1 [cos Θ(k) − 1]2 cos Θ(k) = Re wH 1 (k) g1(k) ||w1(k)|| ||g1(k)|| © AudioLabs 2015 Slide 9 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 11. Constrained IVA (CIVA) IVA with geometric constraints The cost function of IVA with geometric constraints can be expressed as Jciva = Jiva + λ Jp The gradient of Jciva with respect to the elements of the demixing matrix Wciva(k) = E{ϕ(k) (n)y H (n, k)} − W −H (k) Wiva(k) +λ Wp(k) where Wp(k) = C · (cos Θ(k) − 1) g1(k) − w1(k) ||w1(k)||2 Re w H 1 (k)g1(k) © AudioLabs 2015 Slide 10 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 12. Constrained IVA (CIVA) An online CIVA algorithm An online variant of the CIVA algorithm is obtained by omitting the expectation operator in the calculation of Wiva(k). To avoid divergence of the algorithm due to source signal fluctuations, we normalize the gradient matrix at each frame by its Frobenius norm || · ||F Wn(k) = Wn−1(k) − η Wn,civa(k) || Wn,civa(k)||F © AudioLabs 2015 Slide 11 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 13. Content 1. Introduction 2. Constrained IVA (CIVA) 3. Results 4. Conclusions © AudioLabs 2015 Slide 12 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 14. Results: Constrained IVA Experimental setup Dimensions of the simulated room: 7.5m × 5.5m × 3m. Uniform circular array with a diameter of 2.5 cm. Sampling rate 16 kHz, STFT frame 1024 samples (50% overlap). Diffuse noise 30dB SDR, uncorrelated sensor noise 50dB SNR. Segmental speech distortion index (SegSD) and segmental signal to interference ratio (SegSIR) are used as performance measures. © AudioLabs 2015 Slide 13 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 15. Results Fixed number of active sources Proposed constrained IVA algorithm converges in all scenarios. Provides low desired speech signal distortion. T60 = 150 ms T60 = 300 ms segSD segSIR (dB) segSD segSIR (dB) Unprocessed mixture 1.9 1.5 IVA (M = 3, L = 3) 0.21 14.9 0.31 4.7 CIVA (M = 3, L = 3) 0.12 11.7 0.24 5.2 Unprocessed mixture 1.9 1.5 IVA (M = 4, L = 3) 0.27 8.2 0.39 3.8 CIVA (M = 4, L = 3) 0.04 17.0 0.07 8.7 Unprocessed mixture 0.7 -0.3 IVA (M = 4, L = 5) 0.54 0.2 0.57 0.1 CIVA (M = 4, L = 5) 0.05 8.7 0.08 3.6 Table: Performance measures for fixed number of sources © AudioLabs 2015 Slide 14 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 16. Results Time-varying number of active sources Figure: Top: Activity pattern of speech sources over time; Bottom: Segmental SIR improvement over time. M = 6, T60 = 150 ms. © AudioLabs 2015 Slide 15 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 17. Content 1. Introduction 2. Constrained IVA (CIVA) 3. Results 4. Conclusions © AudioLabs 2015 Slide 16 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 18. Conclusions An geometrically constrained online IVA algorithm was proposed to extract the desired speech signal. The applicability of the algorithm to scenarios with fixed and time varying number of interferers was demonstrated. Future work includes comparison of the algorithm against beamforming algorithms and evaluation with recorded data. © AudioLabs 2015 Slide 17 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets
  • 19. Conclusions An geometrically constrained online IVA algorithm was proposed to extract the desired speech signal. The applicability of the algorithm to scenarios with fixed and time varying number of interferers was demonstrated. Future work includes comparison of the algorithm against beamforming algorithms and evaluation with recorded data. Thank you for your attention! © AudioLabs 2015 Slide 17 Constrained Independent Vector Analysis Affan Khan, Maja Taseska, Emanu¨el Habets