Metadata-Version: 2.1
Name: audio-classification-features
Version: 2.0
Summary: Complete Package for Audio Classification
Home-page: https://github.com/Sumit189/Audio-Classification-Using-CNN
Author: Sumit Paul
Author-email: sumit.18.paul@gmail.com
License: UNKNOWN
Keywords: ML,audio-classification,audio-features
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: pandas
Requires-Dist: numpy
Requires-Dist: tqdm
Requires-Dist: scipy
Requires-Dist: librosa
Requires-Dist: keras
Requires-Dist: python-speech-features

# Audio Classification Features
It is made to extract the features from any audio dataset. User's have to provide location of the dataset folder and this library will produce x and y npy files. We also provide custom built Keras model for training.
# Installation
```sh
$ pip install audio_classification_features
```

## Usage
### Making Training Dataset
```py
from audio_classification_features import audio_features as af
af().extractor('dataset')
```

These npy files are loaded with numpy using following commands:
```py
import numpy
x=np.load('x.npy')
y=np.load('y.npy')
```

### Training Custom Built Model
```py
#input_shape generated by this package is of shape (9,13,1)
input_shape=(9,13,1)
```

### Making Predictions
```py 
from audio_classification_features import audio_features as af

#filename example: audio_test.wav
#num_classes same as above 
#model

af().make_prediction(filename,num_classes,model)
```

# Dataset Structure
>folder name of audio will be used as label.

* dataset
    * audio class 1
        * audiofile.wav
    * audio class 2
        * audiofile.wav

# LICENSE
```
GNU General Public License v3.0
```

