Stack Exchange Network

Stack Exchange network consists of 175 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

Questions tagged [audio-recognition]

The tag has no usage guidance.

1
vote
1answer
16 views

Process melspectrograms with convolutional neural network

I am trying to do audio classification with a convolutional neural network. There are six classes. With librosa, I have created melspectrograms for the one second long .wav audio files. It returned ...
0
votes
0answers
7 views

Why is pre and post silence important when collecting speech data?

I've been wondering why it's important, when collecting speech data for AI machine learning, to collect pre and post silence for speech data? Is it just so the machine learns and understands the ...
3
votes
0answers
9 views

looking for databases of audio with labelled 'true' and 'deceptive' sections

For lie-detection research I'm looking for databases of speech audio with labelled 'true' and 'deceptive' sections - does anyone have any leads?
0
votes
1answer
106 views

Accuracy keep changing by changing randomState of classifier

I try to classify car sound samples. Using MLPClassifier from Scikit. I'm getting very different and confusing test results between 2 different test sets, and I am stuck: Training is done with the ...
0
votes
0answers
13 views

Question about signal processing for audio: to use Amplitude_to_db or not? Librosa

So this is my first Audio Processing project (Speaker Recognition), therefore I'm trying to understand a few things. First of all, I understand that there are multiple feature extraction techniques, ...
0
votes
2answers
42 views

Why normalization kills my accuracy

I have a binary car sound classifier. I have a feature set that is extracted from audio with size of 48. I have a model(multi layer neural network) that has around %90 accuracy on test and validation ...
1
vote
1answer
22 views

Audio classification data balance

I'm trying to make a "car sound detector" I have the data from https://urbansounddataset.weebly.com/urbansound.html site. Which has 1000 labeled sounds for 10 different classes(Car sound, dog bark, ...
1
vote
1answer
26 views

Wave form analysis ML algorithm [closed]

I am trying to determine a person's emotions from their speech. This immediately rings Machine Learning bells and the first step in any ML problem is getting and processing data. My first question is, ...
0
votes
2answers
33 views

What is the effect of highly correlated data on a Convolutional Neural Network?

A speech audio sample can be converted to MFCC coefficients for further analysis. I wanted to know the effect of correlated data on a CNN. I know the process of computing the MFCC coefficient, which ...
1
vote
0answers
44 views

Audio signal signal processing for background noise reduction or removal

I am performing simple audio recognition using tensor-flow to spot key word or hot word detection. The graph takes the microphone input directly and performs "Audio spectrometer --> mfcc's --> and ...
2
votes
1answer
51 views

LIME visualization outputs padded regions as important - Mel-spectrogram (audio analysis)

I am encoding audios as Mel-spectrograms and using these Mel-spectrograms as input to my deep learning model (Inception-ResNet V2). The input image is of size 256 X 256, made up of a 128 X 64 ...
1
vote
0answers
18 views

Buy key word audio data set

I am trying to implement audio recognition that can sport key words. I came across google speech dataset but it has a limited data set. Is it possible to buy a new keyword data set ? Can you guys ...
1
vote
2answers
36 views

What is a good method for detection of rare occurencies of speech in noisy audio data?

I have some audio recordings (with relatively static but noisy background, e.g., wind in an open area) with small number of short occurrences of speech (~1% of the total audio duration). What would ...
2
votes
1answer
439 views

How to create speech commands data set

I am planning to create a speech recognition network that recognize few words (voice commands) and came across Speech Commands dataset from google. Apart from available dataset I am planning to add ...
1
vote
0answers
1k views

Audio Analysis : Segment audio based on speaker recognition

I have audio clips of people being interviewed and am trying to split the audio clips using python such that all speech segments of the interviewee are outputted in one audio file (eg .wav format) &...
0
votes
1answer
239 views

LSTM/RNN model fails on new test data - TFLearn

i'm fairly new to ML and at the moment i'm trying to develop a model that can classify spoken digits (0-9) by extracting ...
1
vote
1answer
38 views

How to check if audio samples have only noise or are silent?

I have a wav file I want to split into frames in order to feed it into a machine learning model. The problem is that the audio has silence with some noise at some points. My problem is that I do not ...
1
vote
0answers
58 views

Is machine learning a viable tool to map accent from speech onto text/syllables? [closed]

I'm quite new to coding and even newer to machine learning, so please excuse if this is a stupid question to ask. The project I am thinking about building is a program that helps correct pitch accent (...
-1
votes
0answers
161 views

Audio Analysis: How to identify whether a speaker have finished speaking ?

Just like Siri on Apple phone, I am trying to detect when the speaker has completed speaking. Once I identify that I will convert that to speech to text to understand the question and reply to guy ...
2
votes
1answer
94 views

How to get a feature from sound/audio data learning using machine learning supervised classification?

I'm a bit confused about how data in audio files can be processed for a ML classification model. I have several .wav files that contain dogs barking and cats "meowing". The pipeline is the following: ...
1
vote
2answers
205 views

Working with audio data with different sample rates in Tensorflow

I am trying to implement (as a toy project) some aspects of speech recognition in Tensorflow. The audio files I want to use as training and test data have different sample rates (16, 20, 44 and 44.1 ...
1
vote
1answer
359 views

ESC-50 Audio data for binary classifier

I'm building a binary sound classifier using the ESC-50 dataset. I have taken one class, "dog bark", to be positive and the rest of the 49 classes to be negative. As the dataset is imbalanced I'm ...
2
votes
2answers
270 views

Training an AI to play Starcraft 2 with superhuman level of performance?

I'm interested in working on challenging AI problems, and after reading this article (https://deepmind.com/blog/deepmind-and-blizzard-open-starcraft-ii-ai-research-environment/) by DeepMind and ...
0
votes
1answer
154 views

Downsampling audio files for use in Machine Learning

I'm trying to use the work (Neural Networks) done in this repo: https://github.com/jtkim-kaist/VAD It says this: Note: To apply this toolkit to other speech data, the speech data should be ...
5
votes
1answer
373 views

Detecting voice in a noisy environment

I'm trying to detect voice in a noisy environment. This is a New Zealand suburban/bush soundscape. We're using low quality microphones (but I don't think this has an effect on my problem) to ...
1
vote
0answers
228 views

speech accent recognition data augmentation and training

I am using a Kaggle dataset to learn more about using sound with Deep Learning. I have extracted the mfcc features using this cool library - librosa. However, I want to try feature extraction ...
1
vote
0answers
22 views

Are annotated audio datasets augmented with mutated versions the way image datasets are?

Data augmentation is very standard for annotated image datasets for tasks like image labelling. Images are flipped, rotated, pixelated and so on, to add more training data and make the system robust ...
2
votes
0answers
48 views

Deep Learning models with top-down transfer

Are there Deep Learning models that utilize some sort of top-down information transfer (not only during training)? For example, if the model is used for speech recognition. The lower layers would ...
2
votes
1answer
699 views

(CNN+)RNN-HMM hybrid for learning phonemes from a spectogram

I am currently working on a speech recognition task, on applying deep learning onto the standard acoustic model (gmm-hmm). I've currently generated a spectrogram of my utterances, and using simple ...
1
vote
0answers
419 views

Audio Spectrum Normalization for NeuralNetwork Classification

in my free time I am working on a little machine learning project with focus on NeuralNetworks. It is all about Chord-Classification from an acoustic guitar through a microphone. So here is what I got ...
8
votes
1answer
888 views

CNN for phoneme recognition

I am currently studying this paper, in which CNN is applied for phoneme recognition using visual representation of log mel filter banks, and limited weight sharing scheme. The visualisation of log ...
-1
votes
1answer
183 views

Does the input data representation matter while training CNN for speech recognition?

I am currently doing pattern recognition on spectograms of audio files using convolutional neural networks. The spectograms are made using matplotlib ...
1
vote
1answer
1k views

Input and output feature shapes in CNN for speech recognition

I am currently studying this paper and are trying to understand what exactly the input and output shape is. The paper describes an acoustic model consisting of using cnn-hmm as the acoustic model. ...
5
votes
1answer
2k views

python - What is the format of the WAV file for a Text to Speech Neural Network?

I am creating a Text to Speech system for a phonetic language called "Kannada" and I plan to train it with a Neural Network. The input is a word/phrase while the output is the corresponding audio. ...
5
votes
2answers
1k views

Optimizing CNN network

I am currently trying to recreate the result of this paper, in which they do feature extraction from a "spectogram" of log-melfilter energies.. Since the paper doesn't state what kind of feature I ...
2
votes
1answer
254 views

Dataset: language audio clips and country labels

I'm looking for a dataset containing audio clips of different languages from around the world, including a label indicating which country the clip comes from. Does such a dataset exist? Or is there ...
4
votes
1answer
422 views

Training a CNN with limited weight sharing

I am currently working with speech recognition, in which i would like to try to use CNN instead of the normal feature extraction step. I been reading this paper which proposes method using cnn. The ...
10
votes
1answer
222 views

How does a convolutional ply differ from an ordinary convolutional network?

I am currently working on recreating the results of this paper. In the paper they describe a method for using CNN for features extraction, and have a acoustic model that is Dnn-hmm and pretrained ...
3
votes
2answers
4k views

The effect of an linear layer?

I have the last couple of month worked with an regression problem, of turning a framed audio file into a set of mfcc features, for a speech recognition application I tried a lot different network ...
5
votes
1answer
5k views

Neural network with flexible number of inputs?

Is it possible to create a neural network which provides a consistent output given that the input can be in different length vectors? I am currently in a situation where I have sampled a lot of audio ...
0
votes
1answer
37 views

How to search collection of podcasts (.mp3 files)?

I have a collection of podcasts, in particular the "Talkingmachines" Podcasts ("Human conversation about machine learning"), see http://www.thetalkingmachines.com/blog/ There are 22 episodes, each 1 ...
3
votes
1answer
100 views

Using dhmm_em to form the hmm of mfccs' from song clips

I was working on a project on music genre classification and decided to use an hmm to model my data. After extracting the mfccs' from multiple song clips I get multiple matrices of size 13 * x (...
2
votes
0answers
137 views

Real-time classification of audio - thousands of classes

There is a need to classify an audio stream in real-time - using one from thousands of the defined classes. Up to 5000 classes potentially. What's the best machine learning algorithm for this task? ...
4
votes
2answers
495 views

How to reduce dimensionality of audio data that comes in form of matrices and vectors?

I'm working on a project involved with identifying different types of sounds (such as screams, singing, and bangs) from each other. We've got our data a reasonable number of different transformations ...
11
votes
2answers
3k views

Deep Learning with Spectrograms for sound recognition

I was looking into the possibility to classify sound (for example sounds of animals) using spectrograms. The idea is to use a deep convolutional neural networks to recognize segments in the ...