File size: 3,597 Bytes
94baaee |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 |
# -*- coding: utf-8 -*-
#import zipfile
#zip_ref = zipfile.ZipFile("dataset/train.zip.zip")
#zip_ref.extractall()
#zip_ref.close()
import os
import argparse
import numpy as np
from scipy.io import wavfile
from hmmlearn import hmm
import librosa
import pickle
import numpy as np
import matplotlib.pyplot as plt
from scipy.io import wavfile
from librosa.feature import mfcc
class HMMTrainer(object):
def __init__(self, model_name='GaussianHMM', n_components=4, cov_type='diag', n_iter=1000):
self.model_name = model_name
self.n_components = n_components
self.cov_type = cov_type
self.n_iter = n_iter
self.models = []
if self.model_name == 'GaussianHMM':
self.model = hmm.GaussianHMM(n_components=self.n_components,
covariance_type=self.cov_type, n_iter=self.n_iter)
else:
raise TypeError('Invalid model type')
# X is a 2D numpy array where each row is 13D
def train(self, X):
np.seterr(all='ignore')
self.models.append(self.model.fit(X))
# Run the model on input data
def get_score(self, input_data):
return self.model.score(input_data)
input_folder='aplikasi\dataset'
os.listdir(input_folder)
for dirname in os.listdir(input_folder):
# Get the name of the subfolder
subfolder = os.path.join(input_folder, dirname)
#print(subfolder)
label = subfolder[subfolder.rfind('/') + 1:]
print(label)
sampling_freq, audio = librosa.load("dataset\kata_benda\College_fazrin.wav")
mfcc_features = mfcc(sampling_freq,audio)
print('\nNumber of windows =', mfcc_features.shape[0])
print('Length of each feature =', mfcc_features.shape[1])
mfcc_features = mfcc_features.T
plt.matshow(mfcc_features)
plt.title('MFCC')
hmm_models = []
for dirname in os.listdir(input_folder):
subfolder = os.path.join(input_folder, dirname)
if not os.path.isdir(subfolder):
continue
label = subfolder[subfolder.rfind('/') + 1:]
X = np.array([])
y_words = []
for filename in [x for x in os.listdir(subfolder) if x.endswith('.wav')][:-1]:
filepath = os.path.join(subfolder, filename)
sampling_freq, audio = librosa.load(filepath)
mfcc_features = mfcc(sampling_freq, audio)
if len(X) == 0:
X = mfcc_features[:,:15]
else:
X = np.append(X, mfcc_features[:,:15], axis=0)
y_words.append(label)
print('X.shape =', X.shape)
hmm_trainer = HMMTrainer()
hmm_trainer.train(X)
hmm_models.append((hmm_trainer, label))
hmm_trainer = None
# Test files
input_files = [
'dataset\kata_benda\College_fazrin.wav',
'dataset\kata_kerja\Don_t worry about it_ummu.wav',
'dataset\kata_keterangan\OAF_goose_fear.wav',
'dataset\kata_sifat\Difficult_fazrin.wav'
]
for input_file in input_files:
sampling_freq, audio = librosa.load(input_file)
# Extract MFCC features
mfcc_features = mfcc(sampling_freq, audio)
mfcc_features=mfcc_features[:,:15]
scores=[]
for item in hmm_models:
hmm_model, label = item
score = hmm_model.get_score(mfcc_features)
scores.append(score)
index=np.array(scores).argmax()
print("\nTrue:", input_file[input_file.find('/')+1:input_file.rfind('/')])
print("Predicted:", hmm_models[index][1])
print(f'result : { hmm_models[index][1]}')
# Simpan model dalam bentuk format pkl (pickle)
with open('prolove.pkl', 'wb') as r:
pickle.dump(hmm_models,r) |