Questions tagged [source-separation]

Source separation is the study of separating signal(s) from mixed sources.

Filter by
Sorted by
Tagged with
3
votes
1answer
26 views

Classifying 2 Classes of Ultrasound Signal Using Machine Learning by Frequency Domain

I have two samples which, when exposed to ultrasound, emit their unique frequency responses. As can be seen in the attached figure, where the exciting frequency is 2.25 MHz, sample 1 emits a strong ...
-1
votes
1answer
30 views

Separating a sum of time-shifted signals

Let us suppose that I measure, with an instrument, $h(t,\delta)=f(t)+g(t+\delta)$, where $f(t)$ is the desired signal, and $g(t)$ corresponds to some added undesirable corruption to $h(t,\delta)$. In ...
0
votes
0answers
16 views

Blind source separation for asynchronously observed mixture channels

Given your practical and theoretical expertise: Does ICA work reliably when applied to a multidimensional mixture (observation) $X = (X^1, \cdots, X^d)$ if the different channels $X^i$ of the ...
1
vote
1answer
33 views

Remove a full song from audio recording

I have some audio recordings at 16KHz, which only contain music (BenSound Adventure: https://www.youtube.com/watch?v=0H8JTsG1Jtk). Since I know the music and have a separate wav file, I was wondering ...
2
votes
0answers
151 views

Can machine learning extract two source signals, given a mixed signal?

I have two signals from two sources at a given condition and I have a mixed signal at the same condition that I know is coming from those same two sources. Is it possible to extract the source signals ...
0
votes
0answers
32 views

Realtime Independent Vector Analysis

I have been working on an implementation of Real-time (Online) IVA (independent vector analysis). The paper I am referring to is: Real-Time Independent Vector Analysis for Convolutive Blind Source ...
2
votes
2answers
49 views

Cocktail Party Problem with a Single Signal of Data (Single Mic)

I have been doing some multimodal signal analysis, and sometimes ICA is used for detecting statistically independent components. From my understanding, say if you have 2 sources and 2 receivers/...
0
votes
1answer
47 views

If two signals have different distributions that you know and can model, can you use that to separate them?

For example, if I have a mixed signal composed of signal A added to signal B, and I know that the histogram of signal A is non-Gaussian in the time domain, and that the histogram of signal B is ...
1
vote
0answers
41 views

Signal separation

This is probably a blind signal separation problem of sorts, but it seems like it should be easier than I am finding it. Let’s say I have N time series, each of length [M x 1] that are a superposition ...
0
votes
0answers
23 views

Filtering relationship for sources propagating over a sensor array

A sensor array $y$ measures the superposition of $N$ sources $s$, at time $t$ and position $x$ we have : $$y(t,x)=\sum_{i=1}^{N} s_i(t,x)$$ The sources $s$ travel at constant speed over the array of $...
3
votes
2answers
179 views

Why do we need to estimate eigenvalues?

I am not working in signal processing field, but recently I happen to read a paper which estimates source numbers using Gerschgorin radii, and I feel kind of confused about why we need to estimate ...
0
votes
1answer
61 views

sound source localization - calculation of sound intensity vector angles

I'm working on an acoustic source separation problem where I want to separate the voices of a choir based on intensity vector statistics. I'm currently implementing a paper by Günel called “Acoustic ...
0
votes
0answers
44 views

Is there a repository for source separation test images e.g. the star-galaxy images?

I'm having trouble finding standard images used for testing source-separation algorithms in image processing. For instance, a common example I see is the "star-galaxy" images. Most papers I'...
8
votes
2answers
423 views

Can Principal Component Analysis (PCA) Solve the Cocktail Party Problem?

I'm looking into the cocktail party problem and trying to figure out whether something like Principal Component Analysis is enough to separate out all the various voices at the cocktail party into its ...
1
vote
1answer
41 views

Fast ICA : recover the sources with their mean after prewhitening

I want to apply the FastICA algorithm on a certain dataset. I believe the sources the Fast ICA can recover are able to explain other data as well (in similar way to factor analysis). However, I also ...
3
votes
1answer
169 views

Separating/recovering base signal from two mixed signals, given phase information

I have collected two signals, $B_1(x)$ and $B_2(x)$. The signals start and end at the same $x$-values. Each signal $B_i(x)$ contains: a base signal $b(x)$, which is the same for both, and a signal, ...
-1
votes
1answer
69 views

How to extract two signals from one signal by selecting right index?

Assume that we want to solve $$AX=B$$ Where both $A, X, B$ are matrices. I solved this by using ordinary least squares: $$X = (A^TA)^{-1}A^TB$$ And I got this result for one column in $X$. Here ...
0
votes
1answer
53 views

FastICA: what happens when we have more source than channel?

I am reading the original article proposing FastICA and I have a couple of question on point not covered in the article. Both answering me or providing a source will be very appreciated. I want to ...
0
votes
0answers
38 views

Source separation on multiple observations of the same signal

I possess an ensemble of signal observations: $x_i[n]=s[n]*g_i[n]$, $i=1,2,....,N$ where $N$ is a very large number compared to individual signal lengths (signal lengths are identical). Here, $s[n]$ ...
1
vote
0answers
31 views

What is the best way to separate data using compressive sensing?

In the book Compressed Sensing by Kutyniok et al, the author talks about data separation using sparse representation. In summary, if we have a signal vector $x = x_1 + x_2$ Then, it would be ...
1
vote
1answer
62 views

Separate two measured combinations of two signals with different time delays

I have two signal s1(t) and s2(t) that I want to extract but I can only measure: y1(t) = s1(t) + s2(t-d1) y2(t) = s1(t-d2) + s2(t-d3) The time delays d1, d2 and d3 are unknown, though I have a ...
1
vote
0answers
32 views

Blind source separation from microphone array

I was following this article and I wanted to "borrow" their Idea for my own work. The difference is that I am using a microphone tetrahedral array and not a binaural microphone. I have two different ...
0
votes
0answers
896 views

Splitting an audio signal into multiple audio signals based on frequency range in Librosa

Using the Librosa package in Python, how may I separate an audio signal into multiple audio signals based on frequency range? I have a file music.mp3. I used HPSS ...
0
votes
1answer
70 views

Blind source separation with one source rotated

I have two images: $$ I_1 = w_{11}A + w_{12}B, \\ I_2 = w_{21}A + w_{22} \bar{B}. $$ $A$ and $B$ are unknown. $\bar{B}$ is ${B}$ rotated by 180 degree. For both images, $A$ has higher signal-to-...
1
vote
0answers
31 views

How to classify overlapped signals?

A known signal, signal 1 got overlapped with an unknown signal. Likewise, signal 2 overlapped with another unknown signal. The problem I face now is how to classify the overlapped signals based on the ...
0
votes
0answers
38 views

Incorporate another dimension to non-negative matrix factorization (NMF)?

My first data profile has 3 dimensions; atom pattern (I think it is similar to frequency), time, and signal intensity. I can use supervised NMF for source separation straightforwardly. Because the ...
1
vote
0answers
25 views

Improve NMF for data with partial overlaps in multiple groups?

I want to use NMF to separate true sources from data. My data is in group structure with overlap elements. For example (in the smaller version) group1: contains A,B,C,D,E,F,G patterns group2: ...
4
votes
1answer
348 views

How Is Mixed Norm ($ {L}_{1, 2 }$) Better than $ {L}_{1} $ Norm for Sparse Representation?

Using $ {l}_{1} $-norm regularization for the purpose of achieving sparsity of the solution has been successfully applied a lot. But many times, I found the paper using mixed-norm instead of $l_1$-...
0
votes
4answers
536 views

How to isolate overlapping FM signals?

Analogue radio stations often have a range of frequencies that they can use. Different transmission towers will transmit on different carrier frequencies in this range, because the signal takes time ...
2
votes
2answers
334 views

Audio Steganography: Inaudible Audio Watermarking for Source Identification

What is a relatively inaudible audio watermarking technique to overlay a broadband audio signal on top of playing music without analysing the music for low-bitrate source identification? Ideally ...
0
votes
0answers
153 views

Generalized Sidelobe Cancelling performance

I'm tinkering with different adaptive beam-forming algorithms for a research project in which I want to use a Uniform Rectangular microphones Array (URA) to isolate speech in a room. I am determining ...
1
vote
1answer
39 views

Is sparsity induced penalty in source separation "Entrywise matrix norms"?

I am reading this paper where they introduce norm penalties for source separation. In table 1, the $\log/ l_1$ type is $\sum_{g} log(\epsilon + \lVert H_{g} \rVert_1)$. I wonder this $\lVert H_{g} \...
0
votes
0answers
26 views

Find which groups of ants are coexisted in the location from their signal profiles using NMF

I try to solve biological problem to tell which groups of ants are co-existed at this location by their sounds in specific time frame. My chosen framework is Non-negative Matrix Factorisation (NMF) ...
1
vote
0answers
36 views

Sourse separation from known underdetermined mixing matrix

How to recover uncorrelated N sources from given N-1 signals and known mixing matrix M, (e.g. 9x8 matrix)? If I just use pseudo-inverse matrix M+, my source estimates are correlated with each other ...
0
votes
1answer
124 views

Correlated signals separation with reference

I have a signal S, which needs to be split into two components Sx and Sy. And I have a signal X, which is a reference signal corresponding to Sx. I need to perform this split of S and check that ...
2
votes
0answers
128 views

Speaker diarization tool for acoustic analysis

I'm researching certain acoustic prosodic features during live clinical interactions. Because of the setting, the recordings always have cross channel bleed, and in some cases everything is recorded ...
0
votes
1answer
29 views

Can I start testing my NMF with extension on sample that have only one source?

I am testing my supervised NMF algorithm to extract signal from observation that have only one source inside. I am new here and I wonder this is very weak model or not? Is it acceptable in signal ...
2
votes
1answer
103 views

How to recover signal repeated throughout other signal?

I have an audio signal that contains, at various time offsets, various other short audio signals, often repeated, in addition to other audio content. Here is an example of how that could look like: <...
1
vote
0answers
44 views

What ICA approach is best-suited for semi-blind convolved source separation?

I have a mixed discrete signal $$ A(t_i) = [s(t_i) + a f(t_i+\delta)] \circ g(t_i) $$ for n number of temporal points where $s(t_i)$, $a$, $\delta$ are unknown. $\circ g(t)$ is a convolution, where ...
5
votes
1answer
98 views

Compression Sensing for Blind Source Separation

I am new to Signal Processing, and am interested in compression sensing for audio files. CS is based on the algorithm that, given some sampling of a signal $f$ in order to obtain a smaller (compressed)...
1
vote
0answers
131 views

Best Method to Separate square wave and smooth wave

I have a signal consists of a smooth wave and square waves. Is there comprehensive method can separate square wave and smooth signal? I looked for the likelihood ratio test but cannot find any ...
2
votes
1answer
411 views

Is This Interpretation of Auxiliary Independent Vector Analysis (AuxIVA) Correct?

I am implementing auxiliary function based Independent Vector Analysis (AuxIVA) from Nobutaka Ono's original paper Stable and fast update rules for independent vector analysis based on auxiliary ...
1
vote
0answers
57 views

The effect of the mixed signals probability distribution on the ICA performance

Does the mixed signals distribution affect the ICA performance? I mean the ability of ICA to get the sources. Assume that the mixed signals follows a Gaussian distribution and of course the sources ...
1
vote
1answer
187 views

Extremely large reconstruction errors in NMF

First time in this stack... I am doing some audio analysis. I have a spectrogram ($N\approx 33000, M=1024$) and I need to run an NMF algorithm on it. I am using the Scikit learn implementation. And ...
3
votes
1answer
204 views

Blind Source Separation of Real World Mixture

I have been toying with both non-negative matrix factorization (NMF), independent component analysis (ICA) and independent vector analysis (IVA) for separating speech mixtures. I'm trying to separate ...
8
votes
2answers
2k views

Non-negative matrix factorization for audio separation - why does it work?

Non-Negative Matrix Factorization aims to factorize a matrix $\mathbf V$ into the product of two matrices, $\mathbf V = \mathbf W\mathbf H$, where $\mathbf W$ represents a set of basis vectors and $\...
3
votes
2answers
478 views

Is it correct to subtract two signals acquired over different times and trials to remove the common signal in them?

I have two signals that represent the response of a neuron under two different conditions. Signal 1 (S1): response to Stimulus A Signal 2 (S2): response to Stimulus A+B The response to stimulus A is ...
1
vote
2answers
562 views

Independent component analysis for one observation of a Signal

So I am a complete novice to ICA, so excuse my question if it is bad one, but I have the signal: $$\sin(2\pi x) + \sin(4\pi x) + \textrm{Additive White Gaussian Noise}$$ I want to try to separate ...
1
vote
3answers
282 views

Isolate recurrent pattern vs. evolving background sound

Here is a sound example for what follows. Let's assume we have a signal $$s(t) = r(t) + e(t)$$ where: $r(t)$ is a signal which is recurrent with a given period, i.e. in my example $r(t) = $ the ...
2
votes
0answers
54 views

Supervised instrument recognition

Is it possible to use NMF for a sort of supervised polyphonic instrument recognition? I have training data for a timpani and training data for a bongo (or a bongo, snare and box drum). How can I ...