Spaces:
Runtime error
Runtime error
initial commit
Browse files- .gitattributes +1 -0
- app.py +103 -0
- extract_features.py +33 -0
- finalized_rf.sav +3 -0
.gitattributes
CHANGED
@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
32 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
33 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
34 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
32 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
33 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
34 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
35 |
+
finalized_rf.sav filter=lfs diff=lfs merge=lfs -text
|
app.py
ADDED
@@ -0,0 +1,103 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import wave
|
3 |
+
import matplotlib.pyplot as plt
|
4 |
+
import numpy as np
|
5 |
+
from extract_features import *
|
6 |
+
import pickle
|
7 |
+
import soundfile
|
8 |
+
import librosa
|
9 |
+
|
10 |
+
classifier = pickle.load(open('finalized_rf.sav', 'rb'))
|
11 |
+
|
12 |
+
def emotion_predict(input):
|
13 |
+
input_features = extract_feature(input, mfcc=True, chroma=True, mel=True, contrast=True, tonnetz=True)
|
14 |
+
rf_prediction = classifier.predict(input_features.reshape(1,-1))
|
15 |
+
if rf_prediction == 'happy':
|
16 |
+
return 'Happy π'
|
17 |
+
elif rf_prediction == 'neutral':
|
18 |
+
return 'Neutral π'
|
19 |
+
elif rf_prediction == 'sad':
|
20 |
+
return 'Sad π’'
|
21 |
+
else:
|
22 |
+
return 'Angry π€'
|
23 |
+
|
24 |
+
|
25 |
+
def plot_fig(input):
|
26 |
+
wav = wave.open(input, 'r')
|
27 |
+
|
28 |
+
raw = wav.readframes(-1)
|
29 |
+
raw = np.frombuffer(raw, "int16")
|
30 |
+
sampleRate = wav.getframerate()
|
31 |
+
|
32 |
+
Time = np.linspace(0, len(raw)/sampleRate, num=len(raw))
|
33 |
+
|
34 |
+
fig = plt.figure()
|
35 |
+
|
36 |
+
plt.rcParams["figure.figsize"] = (50,15)
|
37 |
+
|
38 |
+
plt.title("Waveform Of the Audio", fontsize=25)
|
39 |
+
|
40 |
+
plt.xticks(fontsize=15)
|
41 |
+
|
42 |
+
plt.yticks(fontsize=15)
|
43 |
+
|
44 |
+
plt.ylabel("Amplitude", fontsize=25)
|
45 |
+
|
46 |
+
plt.plot(Time, raw, color='red')
|
47 |
+
|
48 |
+
return fig
|
49 |
+
|
50 |
+
|
51 |
+
with gr.Blocks() as app:
|
52 |
+
gr.Markdown(
|
53 |
+
"""
|
54 |
+
# Speech Emotion Detector π΅π
|
55 |
+
This application classifies inputted audio π according to the verbal emotion into four categories:
|
56 |
+
1. Happy π
|
57 |
+
2. Neutral π
|
58 |
+
3. Sad π’
|
59 |
+
4. Angry π€
|
60 |
+
"""
|
61 |
+
)
|
62 |
+
with gr.Tab("Record Audio"):
|
63 |
+
record_input = gr.Audio(source="microphone", type="filepath")
|
64 |
+
|
65 |
+
with gr.Accordion("Audio Visualization", open=False):
|
66 |
+
gr.Markdown(
|
67 |
+
"""
|
68 |
+
### Visualization will work only after Audio has been submitted
|
69 |
+
"""
|
70 |
+
)
|
71 |
+
plot_record = gr.Button("Display Audio Signal")
|
72 |
+
plot_record_c = gr.Plot(label='Waveform Of the Audio')
|
73 |
+
|
74 |
+
record_button = gr.Button("Detect Emotion")
|
75 |
+
record_output = gr.Text(label = 'Emotion Detected')
|
76 |
+
|
77 |
+
with gr.Tab("Upload Audio File"):
|
78 |
+
gr.Markdown(
|
79 |
+
"""
|
80 |
+
## Uploaded Audio should be of .wav format
|
81 |
+
"""
|
82 |
+
)
|
83 |
+
|
84 |
+
upload_input = gr.Audio(type="filepath")
|
85 |
+
|
86 |
+
with gr.Accordion("Audio Visualization", open=False):
|
87 |
+
gr.Markdown(
|
88 |
+
"""
|
89 |
+
### Visualization will work only after Audio has been submitted
|
90 |
+
"""
|
91 |
+
)
|
92 |
+
plot_upload = gr.Button("Display Audio Signal")
|
93 |
+
plot_upload_c = gr.Plot(label='Waveform Of the Audio')
|
94 |
+
|
95 |
+
upload_button = gr.Button("Detect Emotion")
|
96 |
+
upload_output = gr.Text(label = 'Emotion Detected')
|
97 |
+
|
98 |
+
record_button.click(emotion_predict, inputs=record_input, outputs=record_output)
|
99 |
+
upload_button.click(emotion_predict, inputs=upload_input, outputs=upload_output)
|
100 |
+
plot_record.click(plot_fig, inputs=record_input, outputs=plot_record_c)
|
101 |
+
plot_upload.click(plot_fig, inputs=upload_input, outputs=plot_upload_c)
|
102 |
+
|
103 |
+
app.launch()
|
extract_features.py
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import numpy as np
|
2 |
+
import soundfile
|
3 |
+
import librosa
|
4 |
+
|
5 |
+
def extract_feature(file_name, **kwargs):
|
6 |
+
|
7 |
+
chroma = kwargs.get("chroma")
|
8 |
+
contrast = kwargs.get("contrast")
|
9 |
+
mfcc = kwargs.get("mfcc")
|
10 |
+
mel = kwargs.get("mel")
|
11 |
+
tonnetz = kwargs.get("tonnetz")
|
12 |
+
|
13 |
+
with soundfile.SoundFile(file_name) as audio_clip:
|
14 |
+
X = audio_clip.read(dtype="float32")
|
15 |
+
sound_fourier = np.abs(librosa.stft(X)) # Conducting short time fourier transform of audio clip
|
16 |
+
result = np.array([])
|
17 |
+
|
18 |
+
if mfcc:
|
19 |
+
mfccs = np.mean(librosa.feature.mfcc(y=X, sr=audio_clip.samplerate, n_mfcc=40).T, axis=0)
|
20 |
+
result = np.hstack((result, mfccs))
|
21 |
+
if chroma:
|
22 |
+
chroma = np.mean(librosa.feature.chroma_stft(S=sound_fourier, sr=audio_clip.samplerate).T, axis=0)
|
23 |
+
result = np.hstack((result, chroma))
|
24 |
+
if mel:
|
25 |
+
mel = np.mean(librosa.feature.melspectrogram(X, sr=audio_clip.samplerate).T, axis=0)
|
26 |
+
result = np.hstack((result, mel))
|
27 |
+
if contrast:
|
28 |
+
contrast = np.mean(librosa.feature.spectral_contrast(S=sound_fourier, sr=audio_clip.samplerate).T, axis=0)
|
29 |
+
result = np.hstack((result, contrast))
|
30 |
+
if tonnetz:
|
31 |
+
tonnetz = np.mean(librosa.feature.tonnetz(y=librosa.effects.harmonic(X), sr=audio_clip.samplerate).T, axis=0)
|
32 |
+
result = np.hstack((result, tonnetz))
|
33 |
+
return result
|
finalized_rf.sav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:daf37e379f462a4f0e936c39a69aee28e4941c4de46f2e3308711f27042fb514
|
3 |
+
size 3096321
|