Using MFCC's for voice recognition - python

I'm currently using the Fourier transformation in conjunction with Keras for voice recogition (speaker identification). I have heard MFCC is a better option for voice recognition, but I am not sure how to use it.
I am using librosa in python (3) to extract 20 MFCC features. My question is: which MFCC features should I use for speaker identification?
In addition to this I am unsure on how to implement these features. What I would do is to get the necessary features and make one long vector input for a neural network. However, it is also possible to display colors, so could image recognition also be possible, or is this more aimed at speech, and not speaker recognition?
In short, I am unsure where I should start, as I am not very experienced with image recognition and have no idea where to start.
Thanks in advance!!

My question is: which MFCC features should I use for speaker identification?
I shall say that use all of them. Technically MFCC features are output from different filter banks. It is hard to say a priori which of them will be useful.
In addition to this I am unsure on how to implement these features. What I would do is to get the necessary features and make one long vector input for a neural network.
Actually when you extract MFCC for N samples you get an array like N x T x 20 T represents the number of frames in the audio signal after processed for MFCC. I will suggest using Sequence classification with LSTM. This will give better result.

In addition to this I am unsure on how to implement these features.
What I would do is to get the necessary features and make one long
vector input for a neural network.
For each sample, you must have got a 2D matrix of MFCC like N x T X no_mfccs (in your case no_mfccs=20); to make it into one single vector, various researchers take statistics such as mean, var, IQR, etc. to reduce the feature dimension. Some also model it using multivariate regression, and some fit it to a Gaussian mixture model. It depends on the next stage. In your case, you can use statistics to convert into a single vector
OR As told by Parthosarathi, you can use LSTM to preserve sequential information across time frames.
However, it is also possible to display colors, so could image recognition also be possible, or is this more aimed at speech, and not speaker recognition?
I will not recommend you to use spectrogram (image) as a feature vector to neural network because visual images and spectrograms do not accumulate visual objects and sound events information in the same manner.
when you feed image to neural network it assumes that features (pixel values) of an image carry the same meaning regardless of their location. But in case of the spectrogram, location of feature matters a lot.
e.g. Moving the frequencies of a male voice upwards could change its meaning from man to child. Therefore, the spatial invariance that 2D CNN provides might not perform as well for this form of data.
To learn more about it refer: What’s wrong with CNNs and spectrograms for audio processing?

You can use MFCCs with dense layers / multilayer perceptron, but probably a Convolutional Neural Network on the mel-spectrogram will perform better, assuming that you have enough training data.

Related

Is there a time series ML model (like TimesFormer) which extracts features from 4 channel input images?

Regular TimesFormer takes 3 channel input images, while I have 4 channel images (RGBD). I am struggling to find a TimesFormer (or a model similar to TimesFormer) that takes 4 channel input images and extract features from them.
Does anybody know such a model? Preferably, I would like to find pretrained model with weights.
MORE CONTEXT:
I am working with RGBD video frames and have multiclass classification problem at the end. My videos are fairly large, between 2 to 4 minutes so classical time-series models doesn't work for me. So my inputs are RGBD frames/images from the video and at the end I would like to get class prediction.
My idea was to divide the problem into 2 stages:
Extract features from video into smaller dimension with TimesFormer-like model. Result: I would get a new data representation (dataset).
Train clasification ML network with new data to get a class prediction.
As of Jan 2023, I don't think there's any readily available TimeSformer model/code that works on RGBD 4 channel image.
Alternatively, If you are looking for Vision Transformers that can work with depth as well (RGBD data), you have the entire list with state-of-the-art approaches and corresponding code (wherever available) here.
One of the good approach to start with is DepthFormer: Exploiting Long-Range Correlation and Local Information for Accurate Monocular Depth Estimation. You can find the pre-trained models with this approach here.
If you're looking for 3D CNN based object detectors that can work on RGBD data: RGB-D Salient Object Detection via 3D Convolutional Neural Networks is one of the good ones to start with. Code and pre-trained models can be found here.
Since I don't fully understand your problem statement or exact requirement I'm proposed few things that I thought could be helpful.

Can you build a classification model with a simple numpy array

I was wondering if I could build a classification model that just works on simple data I for example stock market classification 1 if it goes up 0 of down. I ask because I never see models like this and tutorials only ever with image or text classification and I typically see regression models working with this form of data
Yes, definitely!
After all image is just a 2D array of numbers.
When you build your classification network, or modify an existing one, you just want to design it so the input layer of the neural network will be a simple array of numbers.
If I have to estimate its performance, I would bet on higher performance relative to a conversational image classifier that does the same task, because every bit of input data is relevant and the data will most likely be smaller without losing any details.

Feature Extraction Using Representation Learning

I'm new to machine learning, and I've been given a task where I'm asked to extract features from a data set with continuous data using representation learning (for example a stacked autoencoder).
Then I'm to combine these extracted features with the original features of the dataset and then use a feature selection technique to determine my final set of features that goes into my prediction model.
Could anyone point me to some resources or demos or sample code of how I could get started on this? I'm very confused on where to begin on this and would love some advice!
Okay, say you have an input of (1000 instances and 30 features). What I would do based on what you told us is:
Train an autoencoder, a neural network that compresses the input and then decompresses it, which has as a target your original input. The compressed representation lies in the latent space and encapsulates information about the input which is not directly accessible by humans. Now you may find such networks in tensorflow or pytorch. Tensorflow is easier and more straightforward so it could be better for you. I will provide this link (https://keras.io/examples/generative/vae/) for a variational autoencoder that may do the job for you. This has Conv2D layers so it performs really well for image data, but you can play around with the architecture. I cannot tell u more because you did not provide more info for your dataset. However, the important thing is the following:
After your autoencoder is trained properly and you need to make sure of it, (it adequately reconstructs the input) then you need to extract the aforementioned latent inputs (you will find more in the link). Now, that will be let's say 16 numbers but you can play with it. These 16 numbers were built to preserve info regarding your input. You said you wanted to combine these numbers with your input so might as well do that and end up with 46 input features. Now the feature selection part has to do with selecting the input features that are more useful for your model. That is not very interesting, you may find more information (https://towardsdatascience.com/feature-selection-techniques-in-machine-learning-with-python-f24e7da3f36e) and one way to select features is by training many models with different feature subsets. Remember, techniques such as PCA are for feature extraction not selection. I cannot provide any demo that does the whole thing but there are sources that can help. Remember, your autoencoder is supposed to return 16 numbers for each training example. Your autoencoder is trained only on your train data, with your train data as targets.

How to classify continuous audio

I have a audio data set and each of them has different length. There are some events in these audios, that I want to train and test but these events are placed randomly, plus the lengths are different, it is really hard to build a machine learning system with using that dataset. I thought fixing a default size of length and build a multilayer NN however, the length's of events are also different. Then I thought about using CNN, like it is used to recognise patterns or multiple humans on an image. The problem for that one is I am really struggling when I try to understand the audio file.
So, my questions, Is there anyone who can give me some tips about building a machine learning system that classifies different types of defined events with training itself on a dataset that has these events randomly(1 data contains more than 1 events and they are different from each other.) and each of them has different lenghts?
I will be so appreciated if anyone helps.
First, you need to annotate your events in the sound streams, i.e. specify bounds and labels for them.
Then, convert your sounds into sequences of feature vectors using signal framing. Typical choices are MFCCs or log-mel filtebank features (the latter corresponds to a spectrogram of a sound). Having done this, you will convert your sounds into sequences of fixed-size feature vectors that can be fed into a classifier. See this. for better explanation.
Since typical sounds have a longer duration than an analysis frame, you probably need to stack several contiguous feature vectors using sliding window and use these stacked frames as input to your NN.
Now you have a) input data and b) annotations for each window of analysis. So, you can try to train a DNN or a CNN or a RNN to predict a sound class for each window. This task is known as spotting. I suggest you to read Sainath, T. N., & Parada, C. (2015). Convolutional Neural Networks for Small-footprint Keyword Spotting. In Proceedings INTERSPEECH (pp. 1478–1482) and to follow its references for more details.
You can use a recurrent neural network (RNN).
https://www.tensorflow.org/versions/r0.12/tutorials/recurrent/index.html
The input data is a sequence and you can put a label in every sample of the time series.
For example a LSTM (a kind of RNN) is available in libraries like tensorflow.

How to visualize features classified by tensorflow?

I'm running the default classify_image code of the imagenet model. Is there any way to visualize the features that it has extracted? If I use 'pool_3:0', that gives me the feature vector. Is there any way to overlay this on top of my image to see which features it has picked as important?
Ross Girshick described one way to visualize what a pooling layer has learned: https://www.cs.berkeley.edu/~rbg/papers/r-cnn-cvpr.pdf
Essentially instead of visualizing features, you find a few images that a neuron fires most on. You repeat that for a few or all neurons from your feature vector. The algorithm needs lots of images to choose from of course, e.g. the test set.
I wrote my implementation of this idea for cifar10 model in Tensorflow today, which I want to share (uses OpenCV): https://gist.github.com/kukuruza/bb640cebefcc550f357c
You could use it if you manage to provide the images tensor for reading images by batches, and the pool_3:0 tensor.

Categories