Fantastic video. Much more helpful than any of the other STFT videos on youtube. Thanks a lot!
@ValerioVelardoTheSoundofAI4 жыл бұрын
Thank you!
@pritamroy7703 жыл бұрын
I cant believe you have only 9k views with this level and clarity of teaching!
@ValerioVelardoTheSoundofAI3 жыл бұрын
Thank you Pritam!
@ice_creamu3 жыл бұрын
I wanted to know what mel spectrograms are but then I watched the first video of the series and now I'm learning a far better-stepped approach to Audio Signal Processing and I'm loving it!
@ashwinalinkil7328 Жыл бұрын
Same!
@A1n2o3s4h5 Жыл бұрын
Us
@LFSDR3 жыл бұрын
Just would thought i would let you know that i am about to finish my thesis on Dolphin vocalization feature extraction and distinction using ML classifiers and just found your videos. Very easy to understand and visualize concepts that took me more time that a wished to understand. Keep up the good work
@islandsounds9357 Жыл бұрын
Thank you so much for this its not only the content and the educational approach is also your style that keeps the interest high
@shravanshenoy3873Ай бұрын
Amazing video. Made complex concepts look so simple
@ramkumarganapathy7795 Жыл бұрын
Thank you so much for all these knowledge sharing. These are one of the best video series I have watched in you-tube!
@janavanrooyen37986 ай бұрын
Fantastic video! It's engaging all the way through and a wonderfully clear explanation of everything.
@shalbbyaali1852Ай бұрын
Thank you so much, most detailed and clear explanation!
@ahnafsamin7464 Жыл бұрын
This tutorial is really helpful! Please keep making contents for us.
@SB-rp8sn Жыл бұрын
Great job on simplifying such a complex topic. Thanks!
@MrOpossumx33 жыл бұрын
The concept of the time/frequency trade off in the STFT is greatly introduced!
@thoaiphan57253 жыл бұрын
Ver comprehensive. Your body language is also awesome! Thanks so much Prof.
@WahranRai4 жыл бұрын
To avoid confusion it will be better to choose another variable (for example L for sfft) instead of N (for fft)
@rangiding993 жыл бұрын
Though I have said that for a couple of times, but still thank you for all this passionate series!
@Waffano2 жыл бұрын
@20:00 How come we don't use a similar definition of frequency bins for DFT (where the frame size is the size of the whole signal ofcourse)?
@hernanvaltierra69123 жыл бұрын
You were born to explain, thank you
@dionsetiawan8798 Жыл бұрын
Cool! This video really helped me for my signal processing subject
@lumpi80611 ай бұрын
Thank your for your great work ! At 23:00 there is something I don't understand : you divide by two the frame size, because of the nyquist rule. You obtain 501. But the 501 frequency bins are for...the interval (0 , sampling rate/2). So, in the end, you divide by 4 ! The sampling rate is divided by 2, THEN the framesize is also divided by 2. Could you explain this,? Thank you.
@radhikasece23743 жыл бұрын
Thanks a lot for the wonderful explanation. I very first started to learn abt MFCC later, am interested in watching all the videos related to the audio signal processing series.
@shanmukhasaratponugupati63083 жыл бұрын
Give this guy a noble prize
@MichaelSievers4 жыл бұрын
Wow, what a great explanation, this has really answered a lot of questions I had about how FFT would work on longer samples. Grazie mille, è stato un piacere guardare il video..
@ValerioVelardoTheSoundofAI4 жыл бұрын
Thank you for your kind words Michael!
@shaidhasan68954 жыл бұрын
Could you please make a video on MFCC?
@ValerioVelardoTheSoundofAI4 жыл бұрын
That is planned over the next few weeks!
@shaidhasan68954 жыл бұрын
@@ValerioVelardoTheSoundofAI waiting for it. And thank you for the whole series!
@canyoupleaserunfast9 ай бұрын
I wish there was Sound Of AI Discord community! Thanks for these videos a lot!
@ValerioVelardoTheSoundofAI9 ай бұрын
There's a TSOAI Slack community though ;)
@chinedueleh30454 жыл бұрын
Wow! I love this series!
@ValerioVelardoTheSoundofAI4 жыл бұрын
Thank you!
@mohammadrahimpoor5132 жыл бұрын
Thank you for your great video. It really helped me understand STFT. I subscribed your channel and will eagerly waiting for your videos.
@Afflictionability Жыл бұрын
lovin your videos man keep up the good work :)
@jennas50392 жыл бұрын
Hi Valerio, may I notice that the calculation for #frames at 22:22 should return 39 and not 19? #frames = (10000 - 1000)/500 + 1 = 39
@ganmohim42732 жыл бұрын
His calculation is correct: (10000 - 1000)/500 + 1 = 19 .🙂
@jennas50392 жыл бұрын
@@ganmohim4273 Ah, yes. Don't know what I was thinking. Apologies
@malikamalika79604 жыл бұрын
Amazing! Love your work and engagement, thank you!
@ValerioVelardoTheSoundofAI4 жыл бұрын
Thank you!
@JIAmitdemwesen2 жыл бұрын
Very informative and well-presented. Thank you!
@cloudhuang7004 жыл бұрын
Thanks for the fantastic video. One quick question. What are the advantage and disadvantage for setting frame_size > window_size ? What is the use case for this parameter choice ?
@mutalasuragemohammed6954 Жыл бұрын
I love the bit; "the k-th frequency at the end temporal uh! bin or n-th frame." 12:18
@mutalasuragemohammed6954 Жыл бұрын
beautifully explained. Thank you
@juleswombat53092 жыл бұрын
Yes this is the dream scenario emerging! - So I am a bit slow but I think I could use e Spectrograms as the input feature layer directly into a convolutional networks or LSTM networks.
@siddharthsharma22482 жыл бұрын
you used a phrase called 'pure term' at 17:03, what do you mean by that?
@Waffano2 жыл бұрын
He said "pure tone"
@adityalesmana21342 жыл бұрын
Awesome explanation, thanks !!!
@stefanhopman91763 жыл бұрын
Thank you for the video!
@proteus510 ай бұрын
When you multiply your signal with the window function in the time domain you are convolving the frequency response of the window function with your signal in the frequency domain. The frequency response of most window functions is some form of a sinc function. Sinc functions are long and ringy, so the result of the convolution is to smear out the frequency response of your output. This reduces the accuracy of the output of the STFT. There are other spectral decomposition algorithms that produce more accurate results. The STFT is popular because of it's ease of computation, not because of it's accuracy.
@DíazRamírezManuel Жыл бұрын
Great work. Thanks you so much
@karennino66392 жыл бұрын
Thank you for sharing!!!
@edsonjunior92672 жыл бұрын
Great job!
@zookaroo21323 жыл бұрын
Very cool lecture !!
@saranshgokhale82983 жыл бұрын
This is great, thanks a lot!
@Moonwalkerrabhi3 жыл бұрын
Awesome explanation !
@Birdsneverfly3 ай бұрын
3 years late, but I love the video
@joshmiller37123 жыл бұрын
Hey man! Your fourier stuff is great! I've been playing with audio a bit and found that tensorflow has an awesome method for getting real-valued spectrograms using a method called MDCT (modified discrete cosine transform). Have you ever considered making a video about that? I'm curious to know how that's different from STFT
@jessicachen92363 жыл бұрын
Is the frequency bin parameter in STFT (in the example is 501 bins) means for each frame in the signal?
@MrDari884 жыл бұрын
Just brilliant once again
@ValerioVelardoTheSoundofAI4 жыл бұрын
Thanks Dario!
@MrDari884 жыл бұрын
@@ValerioVelardoTheSoundofAI I hope you can help me with this doubt. Is the Hann Window applied after STFT to each frequency bin or before the STFT to each sample within the frame? I got a bit confused since in the STFT formula you have used w(n) and on the Hann window formula w(k). Could you please clarify this? Thanks once again for your amazing videos.
@klaimouad7404 жыл бұрын
a Wonderful video, we need to know more about speech processing and especially the mirror process of inverting spectrograms and STFT, could you please suggest me where i can find explanation about the Griffin & Lim algorithm.
@mazmaxman13 жыл бұрын
hello, please if you can provide us the process of the inverse short time fourier transform for overlapped frames, in order to recover the original time domain signal.
@MarineroAndroid2 жыл бұрын
The "temporal" information isn't contained in the DFT phase?, because the DFT is a linear transform, there is not any lose of information, applying the inverse DFT we can get back our signal with it's temporal distribution
@kimamaral30152 ай бұрын
Great Video
@scienceshiritai56044 жыл бұрын
Thank you for the video! If you set 'frame size' bigger than the 'window size', does that increase frequency resolution while keeping the time resolution the same (but at more computational cost)?
@huukhangnguyen34973 жыл бұрын
Wonderful video!!
@goku-np5bk2 жыл бұрын
beautiful!
@burnspeed2 жыл бұрын
Hi @Valerio, this is great stuff. Do you have any recommendations for a book to soak all this in ? These would need intense focus and going multiple back and forth on the videos. Thanks.
@DOMINIK321103 жыл бұрын
You should write your own book, I'd definitely buy it
@rekreator94813 жыл бұрын
Dont you know why when using librosa stft function in python, the resulting number of temporal bins does not equal to math.ceil of (num_of_samples - frame_size) / hop_length + 1, but rather to math.ceil of (num_of_samples) / hop_length? I am processing some audio files and calculating stft for 66150 samples of signal as input to the function, using 2048 window and 512 samples as hop size... So in theory, I should get (66150 - 2048) / 512 + 1 = 126,199... ~127 temporal bins... But rather than that, the temporal output shape of stft does have 130 elements.. How is it calculating the last few windows, for which the function should not actually have enough samples available, as they are out of provided signal range?
@mohammadrezapourtorkan85952 жыл бұрын
I encountered the same thing
@rushrukhrayan10822 жыл бұрын
I encountered the same thing using Librosa. ``` number_of_samples = 661500 FRAME_SIZE = 2048 HOP_LENGTH = 512 debussy_spec = librosa.stft(debussy, n_fft=FRAME_SIZE, hop_length=HOP_SIZE) debussy_spec.shape ``` This gives: (1025, 1292) ~ (#frequency bins, #frames) 1. [According to the given formula] #frames = ((number_of_samples - frame_size) / hop_length) + 1 = ((661500 - 2048) / 512) + 1 = 1288.9921875 = 1289 2. [In reality] #frames = number_of_samples / hop_length = 661500 / 512 = 1291.9921875 = 1292 I feel like, 2 is more intuitive too. We have X number of samples. In each iteration of calculation we move over 512 samples to the right. How many times do we get to do that? Samples/512. Anyone knows where I am going wrong?
@smithflores69682 жыл бұрын
wonderful!!
@simenhex13 жыл бұрын
Thanks for a great series of videos! However, I have a question regarding the resolution trade-off between time and frequency. The time part makes sense, but I do not understand why the frequency resolution depends on the frame size. Obviously something I am missing here, but in my head the frequency range we can represent does not rely on the number of samples, but the sampling rate, ref. the Nyquist sampling theorem. Lets say we have a signal with a sampling rate of 10 samples/sec and we choose a frame size corresponding to 1 second of signal, i.e. 10 samples. Then we can represent frequencies up to max. 5 Hz. If we double the frame size to 2 seconds of signal we now have 20 samples instead of 10. However, the sample rate is still fixed at 10 samples/sec and hence we can still only represent frequencies up to 5Hz...? Would appreciate if you (or anyone else) could explain this.
@JaskaranSingh-hp3zy2 жыл бұрын
I have the same doubt! I think @22:00 he explained that the frequency bins gives the information about the frequencies present in the (0,Sr/2) range equidistant from each other. Bigger the frame size more will be the freq-bins -> more detailed Information about freq when frame size becomes the whole wave -> it will become dft and we will have N samples in the range (0,Sr/2).
@malahatmehraban43402 жыл бұрын
The video was great, thank you. Do you have any instructional videos explaining zero padding?
@ValerioVelardoTheSoundofAI2 жыл бұрын
I've used zero-padding here and there in some videos, but never dedicated a full video to it only.
@malahatmehraban43402 жыл бұрын
@@ValerioVelardoTheSoundofAI Thank you for your quick response. Actually, I'm a master's student in audio signal processing and your videos helped me a lot to do my master's project. Thanks an ocean :)
@zeuspolancosalgado47623 жыл бұрын
You are awesome! I love you!
@petrosgw59283 жыл бұрын
That was a greate video sir. I have got just one question :what happens if we use a very swall hop length for instance 2 or 4?
@ValerioVelardoTheSoundofAI3 жыл бұрын
You would have a lot of redundancy in the data and a greater memory footprint.
@snippletrap3 жыл бұрын
Hop size and frame size are like stride and kernel size in CNNs.
@abdouazizdiop82794 жыл бұрын
Thanks Master
@ValerioVelardoTheSoundofAI4 жыл бұрын
Please call me Valerio :)
@ahaditab63642 жыл бұрын
you are a legend!!!!!!!!
@artyomgevorgyan71674 жыл бұрын
Generally what is the reason for introducing 2 parameters such that one of them reduces the need for the other? I am talking about window size and frame size, in case they are not equal. We could achieve any uniform split just by varying the window size, couldn't we? Just by watching the video, me personally sees no reason for having frame_size != window_size. Of course, I am missing something out, but what?
@ericchuhaochan20664 жыл бұрын
For me, the 2 parameters are conceptually different. Frame size is a param is STFT and window size is a param in window function. Pragmatically, it is pointless to assign frame_size != window_size because those samples in between the gap are going to be 0 padding anyways.
@artyomgevorgyan71674 жыл бұрын
@@ericchuhaochan2066 Agree with you now.
@sarvagyagupta17443 жыл бұрын
Hey, this is an amazing video. Thanks. I have a question though. Through spectrogram, we know the magnitude and phase of the signal at a given time. So is it possible to reconstruct the signal from that domain?
@ValerioVelardoTheSoundofAI3 жыл бұрын
Yes, for reconstruction from a complex spectrogram, you would use the inverse short-time Fourier transform.
@sarvagyagupta17443 жыл бұрын
@@ValerioVelardoTheSoundofAI But spectrogram is the abs of STFT right? So will we need STFT for the reconstruction or just the spectrogram plot will be enough?
@ValerioVelardoTheSoundofAI3 жыл бұрын
@@sarvagyagupta1744 no, the abs of STFT is the magnitude spectrogram, which loses its imaginary part. STFT is complex.
@sarvagyagupta17443 жыл бұрын
@@ValerioVelardoTheSoundofAI but if we take spectogram of STFT, then certain frequencies won't show in the spectogram plot till such time our window doesn't process that, right? So we kinda get some information about the phase of the signal.
@bubblefoil3 жыл бұрын
Thanks! I still don't get the reason for the +1 fft point, though.