Questions tagged [mfcc]

Related to the calculation, verification, usage, and requirements for Mel Frequency Cepstral Coefficients.

Filter by
Sorted by
Tagged with
0
votes
0answers
8 views

Help in understanding MFCC (Mel-Frequency Cepstral Coefficients)

I've been trying to wrap my head around MFCC for a while now, and I have some questions. My understanding is that MFCC is basically windowing, framing, and Fourier Transforming (STFT) the sound signal,...
0
votes
2answers
301 views

MFCC in Speech Recognition and ANN

I have a dataset with over 2000 examples of digits (0 to 9) recorded on wav files. I'm trying to process these files with MFCC so that later I could train an Artificial Neural Network (ANN). My ...
0
votes
1answer
70 views

Audio Activity Detection and Classification

I am starting a new project. Actually, my real intention is to learn speech recognition but for warming up I want to improve in audio signal processing. In my project, I aim to record sounds except ...
4
votes
3answers
3k views

Sinusoidal liftering in implementations of MFCC

Some implementations of MFCC apply sinusoidal liftering as the final step in calculations of MFCC. It is claimed that speech recognition can be significantly improved. For instance, if $\text{MFCC}_i$ ...
0
votes
0answers
24 views

Comparing two audio files of the same sentence and determining if the speaker is the same person

I have a library of audio files where different speakers say the exact same sentence multiple times and I am trying to compare the audio files with each other and determine if the speaker is the same ...
1
vote
0answers
24 views

Get Spectrogram from Filtered Power Spectrum

My objective is to get the higher resolution of spectrogram on the high-frequency area (2000 Hz - 5000 Hz) for a section of speech audio. I know that we typically apply Mel-scaled filter bank to get ...
0
votes
2answers
333 views

can MFCC coefficients alone be used to do voice detection?

Aim of my project is to differentiate between cases of silence, single active speaker and multiple ( simultaneous ) active speakers using as basic ( but effective ) logic as possible. Is it possible ...
1
vote
1answer
36 views

How to compare 2 MFCC (compare them if they where created from same speaker)

There are multiple speakers. Each speaker generates multiple MFCC. If I'm getting two different MFCC, can we know if those MFCC came from 2 different speakers or same speaker ? How can we compare ...
1
vote
0answers
37 views

Logarithmic Amplitude Spectral Subtraction

I'm diving into audio processing and I'm trying to wrap my head around spectral subtraction. I learned there are different approaches do it based on power, magnitude, oversubtraction. My task is to ...
0
votes
1answer
401 views

Machine Learning - any suggestions to solve Python rounding errors?

I am working with Python to isolate voiced segments from music using the Jamendo corpus for singing detection. Training a model, I break my audio into frames, and have a label (0,1) for each frame. ...
0
votes
1answer
950 views

Verification of all steps needed to extract MFCCs from raw audio data

I need some verification on steps required to extract MFCCs from raw audio as I'm finding that some sources outline them slightly differently. Get raw audio frame Run window function (eg Blackman) on ...
0
votes
0answers
34 views

What is the difference between MFCC features and Mel Filter Banks (FBANKs)?

I haven't been able to find any source explaining what the FBANK features are. As a reference, I have been reading this paper here which mentions that they have used 20 MFCC coefficients from 40 mel ...
2
votes
1answer
922 views

Applying DTW on MFCC frames

I want to use MFCC (Mel-frequency cepstral coefficients) and DTW (dynamic time warping) algorithms to make a speech comparison script. I calculated MFCC and got 13 coefficients for each frame. How ...
0
votes
2answers
856 views

Using k Nearest Neighbour & Dynamic Time Warping for keyword spotting

As stated in the title, I am currently researching a method using k-NN, specifically, as either an alternative or as a supplement to the DTW algorithm in keyword spotting based on MFCC. I have read ...
0
votes
1answer
901 views

Comparing band-pass filters and convolutional kernels?

I am currently working on extracting features for speech recognition purposes. I wanted to try an approach similar to MFCC features in which the center frequency of each band pass filters are placed ...
5
votes
2answers
825 views

Dynamic Time Warping - Comparing Values

Ok, so I'm trying to compare two different speech signals and I have come into a problem. Here goes: I have split the signal into blocks, and I have computed the MFCC coefficients of each block. I ...
1
vote
1answer
39 views

how to handle different durations of audio data?

I am new to signal preprocessing, I read about mel_spectrograms, MFCC's. Now I want to apply it and use the CNN model, But the data which I have for practice is having audio of different durations, ...
0
votes
1answer
53 views

Speech Processing applications: stereo data for features

I am working on speech databases that have a stereo format. I want to extract spectral information (Mel-filterbanks, MFCC, LPCC) and also some other prosody features like the fundamental Frequency F0. ...
0
votes
0answers
29 views

How to apply cepstrum coefficient gains (BFCC, MFCC) to an input signal?

I'm trying to understand RNNoise (https://arxiv.org/pdf/1709.08243.pdf, https://github.com/xiph/rnnoise). One aspect in particular, the input signal BFCC coefficients and output signal BFCC ...
0
votes
1answer
20 views

Restricting elemets in mfcc of a signal

I am getting 20-length mfcc of an audio signal. Each element(vector) is of size 5. I am using Librosa. How to give custom size to the each element of mfcc (say 6)?
0
votes
1answer
29 views

Does speech volume affect frequencies?

I have created a little classifier that listens for certain keywords. Sometimes it will not react to a keyword, but if I shout the word, then the classifier will be able to pick it up. The classifier ...
0
votes
0answers
15 views

What acoustics do MFCCs correspond to?

My question is, what frequency ranges do MFCCs correspond to? My line of thinking thus far is that if your highest frequency in the signal was 8000 Hz, you could convert to Mels (= 2834 Mels), take 13 ...
0
votes
0answers
10 views

Sound feature extraction too brief in comparison with manual annotation of sound. How to combine them?

Sound characteristics (features) are taken per e.g. 20 ms, but manual annotation takes place in durations of seconds. If we want to use these characteristics along with labels, how can this be ...
0
votes
1answer
25 views

Using MFCCs for acoustic machine failure prediction

MFCCs are ubiquitously extracted for speech processing tasks, but I would like to know how suitable they are for non-speech processing tasks. Intuitively, it is my understanding that MFCCs are ...
0
votes
0answers
61 views

How do I extract n-dimensional cepstral coefficients with a sliding Hamming window in PyTorch?

I am trying to replicate the features from this paper, described as follows: MFCCs of speech frames are computed using a mel-scale filterbank. We extract 20-dimensional cepstral coefficients with a ...
0
votes
0answers
23 views

In audio analysis, is the first rahmonic from the cepstrum equivalent to the first partial of a pitched sound?

I was reading this text from Oppenheim and Schafer, briefly explaining the concept of the cepstrum. I'm new to this concept. He explains that the cepstrum is useful to quantize the distance between '...
1
vote
0answers
48 views

Understanding MFCC

I am not from electrical or electronic background so my knowledge will be lacking. MFCC is represented by 39 values for each window frame. 12 values are the mel filter-bank and we get 13th value by ...
0
votes
2answers
62 views

Working with a sound's magnitude instead of amplitude

I'm working on a project, where we're recording sound with a piezo-disc which looks a little something like this: Now, unless we're doing something horribly horribly wrong, I've discovered that we're ...
0
votes
0answers
37 views

Compare two spoken words with MFCC and DTW

Processing of the speech signal is done using Matlab. I want to use MFCC (Mel-frequency cepstral coefficients) is implemented using HTK and DTW (dynamic time warping v2.1) algorithms to make a speech ...
0
votes
2answers
57 views

Understanding MFCCs

I am doing research about emotion recognition from speech, by applying machine learning. Most papers are recommending using MFCC features. Therefore, I am currently trying to understand the underlying ...
0
votes
0answers
12 views

Help understand MFCC extraction in SPTK toolkit

The algorithm for MFCC extraction is: Compute the Fourier transform of (a windowed excerpt of) a signal. Map the powers of the spectrum obtained above onto the mel scale, using triangular overlapping ...
0
votes
0answers
143 views

How to save MFCC features and frame energy of several wav files to one .npz file for machine learning training?

I am creating a English-Italian corpus of around 100 hours of English audio from LibriVox aligned with their Italian textual translation. The alignment is at a sentence level and the corpus consist of ...
0
votes
0answers
8 views

Alternatives to MFCCs for Sonar Classification?

Currently I am processing sonar (.wav) files into MFCC features using the librosa library. I use these features as input for sonar classification. Are there any alternatives to MFCCs -- other ...
1
vote
0answers
8 views

Using delta MFCC for signal segmentation

Hello I am focused on implementing a method of speech segmentation into syllables based on delta mfcc. After the calculating of delta, I extracted their maximum local and the next step is to determine ...
0
votes
2answers
32 views

Amplitude of a signal. [-1, 1] vs variants and MFCC implications

I've seen two sort of audio signals, one where the y-axis takes values between -1 and +1. I'd think this means 1 means loudest? Not really sure And the second one I saw goes between $-10 000$ and $+...
0
votes
0answers
69 views

How do I interpret these mfcc graphs?

The first graph represents mfcc plot for female and the second for male. How do I interpret this graph with colors. and what is that color scale on the right? How do I differentiate gender from these ...
0
votes
0answers
53 views

MFCCs: magnitude spectrum vs power spectrum

As I know, the magnitude spectrum and power spectrum have the same information in this step before applying Mel-Filterbank, and the power spectrum may work slightly better in noise. However, I don't ...
0
votes
1answer
71 views

How to train and test deep neural network using MFCC features?

I am working on Voice Disorder Detection problem. I have extracted 13 MFCC features, 13 delta and 13 delta-delta features from each audio file (2 to 4 secs). I extracted these features for each frame (...
0
votes
1answer
150 views

MATLAB: Applying Inverse Fast Fourier Transform to an array of real numbers

I'm trying to understand the difference between applying the Discrete Cosine Transform (DCT) and the Inverse Discrete Fourier Transform (IDFT) to the log Mel-filterbank energies as explained in the ...
0
votes
0answers
299 views

Understanding liftering as the final step in MFCCs features extraction

In the book here, they apply liftering, as a final step of MFCCs features extraction, to isolate the system component by multiplying the whole cepstrum by a rectangular window centred on lower ...
0
votes
2answers
155 views

filterbank: understand the different responses at the center frequency of each filter

I'm computing filterbank by applying 26 triangular filters on a Mel-scale to the power spectrum of an audio frame to extract frequency bands, and I found that some references use filterbank with equal ...
1
vote
1answer
724 views

MFCCs and mean normalization

I'm reading a blog about extracting MFCCs features for Machine Learning applications, but I didn't understand the following points about the mean normalization: To balance the spectrum and improve ...
0
votes
0answers
35 views

Why do we use the term “cepstral” in MFCCs, and how does it related to the cepstral domain?

As I know, we can get the cepstrum by taking the Inverse Discrete Fourier Transform (IDFT) of the log of the power spectrum. And we can get the Mel-frequency cepstral coefficients (MFCCs) by taking ...
0
votes
0answers
16 views

What features capture the semantics of a sentence from a gien audio?

I have been thinking lately and looking up on what could be the possible features that capture the semantics of an audio and came across MFCC: MFCC features represent phonemes (distinct units of sound)...
10
votes
3answers
9k views

Is this a correct interpretation of the DCT step in MFCC calculation?

This is a continuation of the discussion here. I would comment there, but I don't have 50 rep so I'm asking a new question. Here's how I understand the DCT step in the MFCC calculation process: The ...
1
vote
1answer
859 views

Two voice pronunciation comparison similarity MFCC + DTW

Calculate each MFCC to compare wave file A and wave file B, and then use FastDTW to measure the distance after two sets of MFCCs. We compared the four wave files and obtained the Euclidean distance ...
2
votes
1answer
153 views

Calculate the Standard Deviation of Fundamental Frequency (MFCC)

I'm implementing a gunshot detector following the article "Algorithm for Gunshot Detection Using Mel-Frequency Cepstrum Coefficients (MFCC)" (paywall). In the article, the authors uses 22 features ...
1
vote
1answer
75 views

Feature Extraction of Insect Sounds

First of all, forgive me if I sound stupid. I am still learning Audio Processing. So I am trying to make a machine learning model for detecting a specific insect through sound. Here is a sample raw ...
1
vote
0answers
20 views

Alone MFCC can identify difference between sound?

I am doing a project to identify different drum beats. There are lots of features like MFCC, Chroma Values, Spectral Bandwidth, Spectral Centroid, Zero Crossing Rate, etc. But most of the projects are ...
1
vote
1answer
591 views

audio signal reconstruction from MFCC

Transforming an audio signal to Mel Frequency Cepstral Coefficients is broadly used in tasks involving learning on audio. I was wondering, is this transform invertible with some good approximability ...