11
votes
Accepted
What's the correct graphical interpretation of a series of MFCC vectors?
The way MFCC's are always used is by feeding them into the classifier. This can be done on a frame-by-frame basis (12x1 vector), or by concatenating (...
- 11.2k
9
votes
Accepted
Why is each window/frame overlapping?
Why is each window/frame overlapping?
Windowing is a means to stationarize signals. Inside a small enough window, you can expect that the properties of the signal chunk do not vary too fast. And you ...
- 32.3k
9
votes
What are i-vectors and x-vectors in the context of Speech Recognition?
The i-vectors and x-vectors share the ability to represent speech utterance in a compact way (as a vector of fixed size, regardless of length of the utterance).
The extraction algorithms of i-vectors ...
- 231
7
votes
Why is each window/frame overlapping?
More overlap means you end up with more windows (of a given length) per second of audio. More windows (of a given length) requires more FFTs which requires more MACs or FLOPs which generally requires ...
- 35.7k
7
votes
VAD speech databases
Researchers from the Johns Hopkins University have recently released a corpus of music, speech, and noise which, according to them, is suitable for training models for voice activity detection and ...
- 71
7
votes
What is the explanation that filter EQ changes "Ah" to "O"?
Expanding a bit on Tim Wescott's answer:
Vowels are made by a combination of an excitation and a resonant filter. The excitation comes from the glottis in your throat. The speed of that vibration ...
- 48.2k
6
votes
How does Siri recognize me saying "Hey Siri"?
"Ok Google" is described in many publications by Google
Automatic Gain Control and Multi-style Training for Robust Small-Footprint Keyword Spotting with Deep Neural Networks
Convolutional Neural ...
- 1,119
4
votes
Cepstrum calculus disambiguation
A common technique for computing the inverse fft is to invert the imaginary part of the input array, perform a forward fft and then invert the imaginary part of the output array. In the case of the ...
- 666
4
votes
Accepted
What is the explanation that filter EQ changes "Ah" to "O"?
What is the explanation of this phenomenon?
Your vocal cords have a very raspy sort of vibration -- they generate spectral components at the fundamental, and at many many harmonics going up from ...
- 13.3k
3
votes
VAD speech databases
Free VAD Data set recorded in real environment with ground truth label:
refer: https://github.com/jtkim-kaist/VAD-Toolkit
- 41
3
votes
Remove a known wav file from recorded file
Assuming that there are no non-linear effects between the sound N source and the microphone (such as AGC), you might try to estimate the impulse response of the channel between the source for sound N ...
- 35.7k
3
votes
Accepted
How to edit audios so that they have the same length
ImageMagick is a great tool to edit multiple images in the command line. Thanks to your question, I just discovered SoX, or Sound eXchange:
SoX is a cross-platform (Windows, Linux, MacOS X, etc.) ...
- 32.3k
3
votes
Is there any way that we can perform speech recognition without using Fourier transforms?
All recognition tasks (doesn't even have to be speech recognition) are reductions of a very high-dimensional signal (your speech recording's dimension is the number of audio samples!) to a low-...
- 32.5k
3
votes
Accepted
filterbank: understand the different responses at the center frequency of each filter
Normalizing the filterbanks by their widths is optional and totally up to you (similarly to the warping scale Mel/Bark). Depending on your application, you can start without normalization and see what ...
- 11.2k
2
votes
Matching LPC frequency response to raw sample data frequency response
The computed LPC coefficients are the weights $a_i$, which predict a future signal value based on $p$ previous samples of the input signal:
$$\hat{x}[n] = \sum_{i=1}^{p} a_i x[n-i],$$
where $\hat{x}[n]...
- 304
2
votes
Compare two audio tracks
You should think about what level it makes sense to "compare".
You can look at your signals as time-series and compute statistics (e.g. cross-correlation, phase-correlation, etc). Alternatively, ...
- 1,770
2
votes
How should this fft output be intepreted?
you can definitely extract the amplitude for a particular frequency bin(which might include few frequencies depending on your DFT resolution).
Am I missing something? or am I right when i say ...
- 687
2
votes
Accepted
How should this fft output be intepreted?
"[...] the real part is the amplitude, and the complex part is the phase [...]" No, complex numbers (and that's what the output of a DFT/FFT is) can be represented either by their real and imaginary ...
- 92.4k
2
votes
Accepted
The minimal signal to noise ratio (SNR) for people to understand a speech in the noisy background
The Equi-loudness contours, chart provided at your link, is for hearing thresholds for isolated tones. Assuming speech is a sum of tones is a wrong assumption. But even if we go ahead with this ...
- 393
2
votes
Accepted
Applying dtw on mfcc
Let's unpack this code:
1 mfcc1=mfcc1';
2 mfcc2=mfcc2';
3 M=simmx(mfcc1,mfcc2);
4 [p,q,c]=dp(1-M);
5 v=c(size(c,1),size(c,2))
in line 3, ...
- 1,770
2
votes
Deep learning model for phone recognition - issues with dimension the model
Since next layer is fully connected it does not really matter what shape your pooling output would be. You have 14x100, you can rearrange them as 1x1400 as input for next layer, 1000 elements as ...
- 1,119
2
votes
Accepted
How does noise reduction for speech recognition differ from noise reduction that is supposed to make speech more "intelligible" for humans?
I don't really find papers that discuss this difference.
There are whole books on the subject:
Robust Automatic Speech Recognition 1st Edition
Do speech intelligibility and speech quality ...
- 1,119
2
votes
Speech recognition question
I didn't hear about free software that could transcribe audio to text but there are some free online solutions. For example Speechnotes will translate speech recorded over a microphone to text. The ...
- 369
2
votes
Why is LPCC and LPC used in speech recognition
I think a high-level explanation is the right thing to give here:
For recognition of a producing process (in this case: speech) from a signal (in this case: audio), you need to find a mapping from ...
- 32.5k
2
votes
Remove a known wav file from recorded file
This is not just a channel estimation problem. Because the known reference signal and the recording have not been clock-synchronised you will have a small amount of temporal drift between the clocks. ...
- 4,603
2
votes
Remove a known wav file from recorded file
Description of Source Separation Algorithm
The approach that I would take is based on a Semi-supervised Non-negative Matrix Factorization. This would work assuming that:
The audio file that you are ...
- 11.2k
2
votes
How to edit audios so that they have the same length
In order to make your audio wav file longer you can use sox in linux :
sox input.wav output.wav trim 0 00:01
This will take 1s of the audio input and put it in output.wav. If it's shorter than 1s ...
- 121
2
votes
Accepted
What Is the Point of Doing the Zero Padding?
You may think of it as efficient way to apply Dirichlet Window Based interpolation in the Fourier Domain.
The advantage of applying the interpolation using Zero Padding in the Time Domain is very ...
- 20.5k
2
votes
MFCCs and mean normalization
Why did he add the epsilon value 1e-8
This is so you don't end up taking log of zero later, although since the first operation isn't guaranteed to be >=0, I'm not sure if it's necessary or helpful
- 121
Only top scored, non community-wiki answers of a minimum length are eligible