feat: Audio Comparison - Signal Framing Completed

BE-Pronunciation/Audio Comparison.py

@@ -51,99 +51,61 @@ np.allclose(y_filt, np.concatenate([y_filt_1, y_filt_2]))
 
 # Framing and windowing of voice signals
 
-# In[21]:
+# In[6]:
 
 
-
- def framing(sig, fs=16000, win_len=0.025, win_hop=0.01):
-     """
-     transform a signal into a series of overlapping frames.
-
-     Args:
-         sig            (array) : a mono audio signal (Nx1) from which to compute features.
-         fs               (int) : the sampling frequency of the signal we are working with.
-                                  Default is 16000.
-         win_len        (float) : window length in sec.
-                                  Default is 0.025.
-         win_hop        (float) : step between successive windows in sec.
-                                  Default is 0.01.
-
-     Returns:
-         array of frames.
-         frame length.
-     """
-
-
-# 
-#     # pre-emphasis
-#     if pre_emph:
-#         sig = pre_emphasis(sig=sig, pre_emph_coeff=0.97)
-#  
-#     # -> framing
-#     frames, frame_length = framing(sig=sig,
-#                                    fs=fs,
-#                                    win_len=win_len,
-#                                    win_hop=win_hop)
-#  
-#     # -> windowing
-#     windows = windowing(frames=frames,
-#                         frame_len=frame_length,
-#                         win_type=win_type)
-#  
-
-# In[25]:
-
-
-# compute frame length and frame step (convert from seconds to samples)
+import numpy as np
 
 
-frame_length = win_len * fs
-frame_step = win_hop * fs
-signal_length = len(sig)
-frames_overlap = frame_length - frame_step
+def framing(sig, fs=16000, win_len=0.025, win_hop=0.01):
+    """
+    transform a signal into a series of overlapping frames.
+
+    Args:
+        sig            (array) : a mono audio signal (Nx1) from which to compute features.
+        fs               (int) : the sampling frequency of the signal we are working with.
+                                 Default is 16000.
+        win_len        (float) : window length in sec.
+                                 Default is 0.025.
+        win_hop        (float) : step between successive windows in sec.
+                                 Default is 0.01.
+
+    Returns:
+        array of frames.
+        frame length.
+    """
+    # compute frame length and frame step (convert from seconds to samples)
+    frame_length = win_len * fs
+    frame_step = win_hop * fs
+    signal_length = len(sig)
+    frames_overlap = frame_length - frame_step
 
     # Make sure that we have at least 1 frame+
-
-num_frames = np.abs(signal_length - frames_overlap) // np.abs(frame_length - frames_overlap)
-rest_samples = np.abs(signal_length - frames_overlap) % np.abs(frame_length - frames_overlap)
-
-     # Pad Signal to make sure that all frames have equal number of samples
-     # without truncating any samples from the original signal
-        
-if rest_samples != 0:
-    pad_signal_length = int(frame_step - rest_samples)
-    z = np.zeros((pad_signal_length))
-    pad_signal = np.append(sig, z)
-    num_frames += 1
-else:
-    pad_signal = sig
-
-     # make sure to use integers as indices
-        
-frame_length = int(frame_length)
-frame_step = int(frame_step)
-num_frames = int(num_frames)
-
-     # compute indices
-    
-idx1 = np.tile(np.arange(0, frame_length), (num_frames, 1))
-idx2 = np.tile(np.arange(0, num_frames * frame_step, frame_step),
-            (frame_length, 1)).T
-indices = idx1 + idx2
-frames = pad_signal[indices.astype(np.int32, copy=False)]
-return frames
-
-
-# In[ ]:
-
-
-
-
-
-# In[ ]:
-
-
-
+    num_frames = np.abs(signal_length - frames_overlap) // np.abs(frame_length - frames_overlap)
+    rest_samples = np.abs(signal_length - frames_overlap) % np.abs(frame_length - frames_overlap)
+
+    # Pad Signal to make sure that all frames have equal number of samples
+    # without truncating any samples from the original signal
+    if rest_samples != 0:
+        pad_signal_length = int(frame_step - rest_samples)
+        z = np.zeros((pad_signal_length))
+        pad_signal = np.append(sig, z)
+        num_frames += 1
+    else:
+        pad_signal = sig
+
+    # make sure to use integers as indices
+    frame_length = int(frame_length)
+    frame_step = int(frame_step)
+    num_frames = int(num_frames)
+
+    # compute indices
+    idx1 = np.tile(np.arange(0, frame_length), (num_frames, 1))
+    idx2 = np.tile(np.arange(0, num_frames * frame_step, frame_step),
+                   (frame_length, 1)).T
+    indices = idx1 + idx2
+    frames = pad_signal[indices.astype(np.int32, copy=False)]
+    return frames
 
 
 # In[ ]: