Spaces:
Runtime error
Runtime error
chidojawbreaker
commited on
Commit
•
e1f22dd
1
Parent(s):
d4d5ba9
Update app.py
Browse files
app.py
CHANGED
@@ -1,13 +1,26 @@
|
|
1 |
import gradio as gr
|
2 |
import numpy as np
|
|
|
3 |
import joblib
|
4 |
import librosa
|
5 |
from sklearn.preprocessing import StandardScaler
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
|
7 |
model = joblib.load('UTI.pkl')
|
8 |
|
9 |
def predictor(audio_filename):
|
10 |
-
y,sr = librosa.load(audio_filename,mono=True,duration=5)
|
11 |
|
12 |
chroma_stft = np.mean(librosa.feature.chroma_stft(y=y, sr=sr))
|
13 |
rmse = np.mean(librosa.feature.rms(y=y))
|
@@ -41,17 +54,16 @@ def predictor(audio_filename):
|
|
41 |
mfcc20 = v[19]
|
42 |
|
43 |
features = np.array([[chroma_stft,rmse,spec_cent,spec_bw,rolloff,zcr,mfcc1,mfcc2,mfcc3,mfcc4,mfcc5,mfcc6,mfcc7,mfcc8,mfcc9,mfcc10,mfcc11,mfcc12,mfcc13,mfcc14,mfcc15,mfcc16,mfcc17,mfcc18,mfcc19,mfcc20]])
|
44 |
-
|
45 |
-
prediction = model.predict(StandardScaler().fit_transform(features))
|
46 |
|
47 |
-
|
|
|
|
|
48 |
result = 'Normal'
|
49 |
else:
|
50 |
result = 'Infected'
|
51 |
return result
|
52 |
|
53 |
-
|
54 |
app = gr.Interface(fn=predictor,
|
55 |
inputs=gr.Audio(source="upload",type="filepath",label="Please Upload Audio file here:"),
|
56 |
-
outputs=gr.Textbox(label="Result"),title="SMART
|
57 |
-
app.launch()
|
|
|
1 |
import gradio as gr
|
2 |
import numpy as np
|
3 |
+
import pandas as pd
|
4 |
import joblib
|
5 |
import librosa
|
6 |
from sklearn.preprocessing import StandardScaler
|
7 |
+
from sklearn import preprocessing
|
8 |
+
from sklearn.model_selection import train_test_split
|
9 |
+
|
10 |
+
dataset = pd.read_csv('UTIv2.csv')
|
11 |
+
dataset = dataset.drop('filename',axis=1)
|
12 |
+
x = dataset.iloc[:, :-1].values
|
13 |
+
y = dataset.iloc[:, -1].values
|
14 |
+
encoder = preprocessing.LabelEncoder()
|
15 |
+
y = encoder.fit_transform(y)
|
16 |
+
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size = 0.2, random_state = 0)
|
17 |
+
sc = StandardScaler()
|
18 |
+
sc = sc.fit(x_train)
|
19 |
|
20 |
model = joblib.load('UTI.pkl')
|
21 |
|
22 |
def predictor(audio_filename):
|
23 |
+
y, sr = librosa.load(audio_filename, mono=True, duration=5)
|
24 |
|
25 |
chroma_stft = np.mean(librosa.feature.chroma_stft(y=y, sr=sr))
|
26 |
rmse = np.mean(librosa.feature.rms(y=y))
|
|
|
54 |
mfcc20 = v[19]
|
55 |
|
56 |
features = np.array([[chroma_stft,rmse,spec_cent,spec_bw,rolloff,zcr,mfcc1,mfcc2,mfcc3,mfcc4,mfcc5,mfcc6,mfcc7,mfcc8,mfcc9,mfcc10,mfcc11,mfcc12,mfcc13,mfcc14,mfcc15,mfcc16,mfcc17,mfcc18,mfcc19,mfcc20]])
|
|
|
|
|
57 |
|
58 |
+
prediction = model.predict(sc.transform(features))
|
59 |
+
|
60 |
+
if prediction[0] == 1:
|
61 |
result = 'Normal'
|
62 |
else:
|
63 |
result = 'Infected'
|
64 |
return result
|
65 |
|
|
|
66 |
app = gr.Interface(fn=predictor,
|
67 |
inputs=gr.Audio(source="upload",type="filepath",label="Please Upload Audio file here:"),
|
68 |
+
outputs=gr.Textbox(label="Result"),title="SMART LUTS DETECTOR",description="UTI Prediction Model",examples=[["normal 1_rn.wav"]])
|
69 |
+
app.launch()
|