deepugaur commited on
Commit
5b6f753
1 Parent(s): 6b708e3

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +21 -25
app.py CHANGED
@@ -1,44 +1,40 @@
1
- pip install -r requirements.txt
2
-
3
- from datetime import datetime
4
  import tensorflow as tf
5
- from tensorflow.keras.models import load_model
6
  import numpy as np
7
  import librosa
 
8
 
9
- # Load pre-trained models
10
- speech_to_text_model = load_model('speech_to_text_model.h5')
11
- translation_model = load_model('translation_model.h5')
12
 
13
- def preprocess_audio(file_path):
14
- # Load and preprocess the audio file
15
- audio, sr = librosa.load(file_path, sr=16000)
16
  mfccs = librosa.feature.mfcc(y=audio, sr=sr, n_mfcc=13)
17
  return np.expand_dims(mfccs, axis=0)
18
 
19
  def translate_speech_to_text(audio_file):
20
- # Preprocess the audio file
21
  audio_features = preprocess_audio(audio_file)
22
-
23
- # Predict text from audio
24
  predicted_text = speech_to_text_model.predict(audio_features)
25
-
26
- # Translate text
27
  translated_text = translation_model.predict([predicted_text])
28
-
29
  return translated_text
30
 
31
  def is_after_six_pm():
32
  current_time = datetime.now()
33
  return current_time.hour >= 18
34
 
35
- def main(audio_file):
36
- if is_after_six_pm():
37
- translated_text = translate_speech_to_text(audio_file)
38
- print("Translated Text:", translated_text)
39
- else:
40
- print("Service available only after 6 PM IST.")
 
 
 
 
 
 
41
 
42
- # Example usage
43
- audio_file_path = 'path/to/your/audiofile.wav'
44
- main(audio_file_path)
 
1
+ import streamlit as st
 
 
2
  import tensorflow as tf
 
3
  import numpy as np
4
  import librosa
5
+ from datetime import datetime
6
 
7
+ # Load models
8
+ speech_to_text_model = tf.keras.models.load_model('speech_to_text_model.h5')
9
+ translation_model = tf.keras.models.load_model('translation_model.h5')
10
 
11
+ def preprocess_audio(file):
12
+ audio, sr = librosa.load(file, sr=16000)
 
13
  mfccs = librosa.feature.mfcc(y=audio, sr=sr, n_mfcc=13)
14
  return np.expand_dims(mfccs, axis=0)
15
 
16
  def translate_speech_to_text(audio_file):
 
17
  audio_features = preprocess_audio(audio_file)
 
 
18
  predicted_text = speech_to_text_model.predict(audio_features)
 
 
19
  translated_text = translation_model.predict([predicted_text])
 
20
  return translated_text
21
 
22
  def is_after_six_pm():
23
  current_time = datetime.now()
24
  return current_time.hour >= 18
25
 
26
+ def main():
27
+ st.title("Audio Translation App")
28
+
29
+ uploaded_file = st.file_uploader("Choose an audio file", type="wav")
30
+
31
+ if uploaded_file is not None:
32
+ if is_after_six_pm():
33
+ st.write("Processing...")
34
+ translated_text = translate_speech_to_text(uploaded_file)
35
+ st.write("Translated Text:", translated_text)
36
+ else:
37
+ st.write("Service available only after 6 PM IST.")
38
 
39
+ if __name__ == "__main__":
40
+ main()