Signal Processing Stack Exchange
Signal Processing Stack Exchange

Questions tagged [stft]

Short-Time Fourier Transform

Short-Time Fourier Transform (STFT) is a time-frequency analysis tool that repeatedly applies the Discrete Fourier Transform (DFT or FFT) to windowings of short snippets of a longer signal. It is equivalently convolutions with windowed complex sinusoids, or bandpass filtering, where the "hop" parameter is the stride or subsampling factor. It's a function of two independent variables: frequency in the DFT, and time of the center of the sliding window.

314 questions
8
votes
3 answers

How critical is the selection of the window function in STFTs?

I have a sum of periodic signals that I am trying to untangle using time-frequency analysis. I seem to get wildly different results depending on the window length and shape. This is a problem because I want to develop an automated, and hopefully…
Emre
  • 2,877
  • 17
  • 22
5
votes
3 answers

Is "Windowed Fourier Transform" a synonym for "STFT"?

Sometimes I find the notation WFT (Windowed Fourier Transform) while other times I see STFT (Short Time Fourier Transform). Are they the same?
Diego Sacconi
  • 53
  • 4
4
votes
1 answer

What is first order difference?

I don't understand what is a first order sequence. For more context: I am reading a paper where the author extracts different sequences (in the STFT domain) and then "The resulting sequences are differentiated (via first-order difference)". What is…
KM158
  • 41
  • 1
  • 2
1
vote
1 answer

Filterbank Interpretation of Inverse STFT With Hop

The slide below, borrowed from CMU, shows the typical interpretation of the "filterbank interpretation" of signal reconstruction from a Short Time Fourier Transform. As far as I can tell, the hop-size here is 1, i.e., there is no frame decimation…
Novak
  • 175
  • 1
  • 9
1
vote
2 answers

Short-time Fourier transform Tradeoffs

In the Short-time Fourier transform what are the tradeoffs between different: window functions window lengths sample rates
Tom Huntington
  • 376
  • 3
  • 18
1
vote
0 answers

Scipy stft, how is shape of f determined?

Using scipy stft, like this : sample_rate, samples = scipy.io.wavfile.read(sample_dir/'original'/test_sample_name) f, t, Zxx = scipy.signal.stft(samples.transpose(), sample_rate) f.shape is equal to 129. My understanding is that this number…
Statistic Dean
  • 151
  • 1
  • 6
0
votes
1 answer

Is there a FFT algorithm with the circular buffering?

I have an audio signal and i want to process it sequentially. Is there an algorithm which allows to modify the spectrum of a signal when i remove a part of signal and add the other one? For example, i obtain a set of 256 samples of input data. Then…
John
  • 61
  • 5
0
votes
0 answers

Clustering real and imaginary values within a frequency row of an STFT

I am analyzing a short audio .wav file of a drum and piano for source separation. (not using interpretable languages like Librosa, Scikit, Matlab, or R, but rather a compilable language in .NET). When considering the hierarchical clustering…
user16354
-1
votes
1 answer

The effect of doubling the sample rate in Spectrogram

If we have an audio file and we only change the sample rate (double it), what changes our spectrogram will have (after using the stft)?
Kurosh Ghanizadeh
  • 99
  • 1
-1
votes
1 answer

Overlap frames in short term fourier transform

I'm trying to understand the aim of overlapping frames in the short term fourier transform. Why is it important to do it? I have been told that it's to avoid "problems" at the frontiers of frames. But does anyone have a more precise or mathematical…
user3408121
  • 101