And its done! MD file needs writing.

.gitignore

@@ -14,6 +14,8 @@ outputs/metrics/*
 outputs/states/*
 !outputs/states/.dummy.txt
 
+*.wav
+
 # Created by .ignore support plugin (hsz.mobi)
 ### Python template
 # Byte-compiled / optimized / DLL files

helpers/data_feeder.py

@@ -89,7 +89,7 @@ def epoch_it():
 
 
 def data_feeder_testing(window_size, fft_size, hop_size, seq_length, context_length,
-                        batch_size, debug):
+                        batch_size, debug, sources_list=None):
     """Provides an iterator over the testing examples.
 
     :param window_size: The window size to be used for the time-frequency transformation.
@@ -106,11 +106,17 @@ def data_feeder_testing(window_size, fft_size, hop_size, seq_length, context_len
     :type batch_size: int
     :param debug: A flag to indicate debug
     :type debug: bool
+    :param sources_list: The file list provided for using the MaD-TwinNet.
+    :type sources_list: list[str]
     :return: An iterator that will provide the input and target values.\
              The iterator yields (mix, mix magnitude, mix phase, voice true, bg true) values.
     :rtype: callable
     """
-    sources_list = _get_files_lists('testing')[-1]
+    if sources_list is None:
+        usage_case = False
+        sources_list = _get_files_lists('testing')[-1]
+    else:
+        usage_case = True
     hamming_window = hamming(window_size, True)
 
     def testing_it():
@@ -120,7 +126,7 @@ def testing_it():
                 sources_parent_path=sources_list[index],
                 window_values=hamming_window, fft_size=fft_size, hop=hop_size,
                 seq_length=seq_length, context_length=context_length,
-                batch_size=batch_size
+                batch_size=batch_size, usage_case=usage_case
             )
 
             if debug:
@@ -130,7 +136,8 @@ def testing_it():
 
 
 def data_process_results_testing(index, voice_true, bg_true, voice_predicted,
-                                 window_size, mix, mix_magnitude, mix_phase, hop, context_length):
+                                 window_size, mix, mix_magnitude, mix_phase, hop,
+                                 context_length, output_file_name=None):
     """Calculates SDR and SIR and creates the resulting audio files.
 
     :param index: The index of the current source/track.
@@ -153,6 +160,11 @@ def data_process_results_testing(index, voice_true, bg_true, voice_predicted,
     :type hop: int
     :param context_length: The context length in frames.
     :type context_length: int
+    :param output_file_name: The output file name for the predicted voice\
+                             and background music. If this argument is not
+                             None, then the function just synthesizes the
+                             voice and the background music, and saves them.
+    :type output_file_name: list[str] | None
     :return: The values of SDR and SIR for each of the frames in\
              the current track, for both voice and background music.
     :rtype: (list[numpy.core.multiarray.ndarray], list[numpy.core.multiarray.ndarray])
@@ -164,29 +176,42 @@ def data_process_results_testing(index, voice_true, bg_true, voice_predicted,
 
     # Removing the samples that no estimation exists
     mix = mix[context_length * hop:]
-    voice_true = voice_true[context_length * hop:]
-    bg_true = bg_true[context_length * hop:]
-
-    min_len = min(len(voice_true), len(voice_hat))
+    if output_file_name is None:
+        voice_true = voice_true[context_length * hop:]
+        bg_true = bg_true[context_length * hop:]
+        min_len = min(len(voice_true), len(voice_hat))
+        example_index = index + 1
+    else:
+        voice_true = None
+        bg_true = None
+        example_index = None
+        min_len = min(len(mix), len(voice_hat))
 
     # Background music estimation
     bg_hat = mix[:min_len] - voice_hat[:min_len]
 
-    example_index = index + 1
+    if output_file_name is None:
+        voice_hat_path = output_audio_paths['voice_predicted'].format(p=example_index)
+        bg_hat_path = output_audio_paths['bg_predicted'].format(p=example_index)
+        wav_write(voice_true, file_name=output_audio_paths['voice_true'].format(p=example_index), **wav_quality)
+        wav_write(bg_true, file_name=output_audio_paths['bg_true'].format(p=example_index), **wav_quality)
+        wav_write(mix, file_name=output_audio_paths['mix'].format(p=example_index), **wav_quality)
 
-    wav_write(voice_true, file_name=output_audio_paths['voice_true'].format(p=example_index), **wav_quality)
-    wav_write(voice_hat, file_name=output_audio_paths['voice_predicted'].format(p=example_index), **wav_quality)
+        # Metrics calculation
+        sdr, sir = _get_me_sdr_and_sir(bss_eval.bss_eval_images_framewise(
+            [voice_true[:min_len], bg_true[:min_len]],
+            [voice_hat[:min_len], bg_hat[:min_len]]
+        ))
 
-    wav_write(bg_true, file_name=output_audio_paths['bg_true'].format(p=example_index), **wav_quality)
-    wav_write(bg_hat, file_name=output_audio_paths['bg_predicted'].format(p=example_index), **wav_quality)
+    else:
+        voice_hat_path = output_file_name[0]
+        bg_hat_path = output_file_name[1]
 
-    wav_write(mix, file_name=output_audio_paths['mix'].format(p=example_index), **wav_quality)
+        sdr = None
+        sir = None
 
-    # Metrics calculation
-    sdr, sir = _get_me_sdr_and_sir(bss_eval.bss_eval_images_framewise(
-        [voice_true[:min_len], bg_true[:min_len]],
-        [voice_hat[:min_len], bg_hat[:min_len]]
-    ))
+    wav_write(voice_hat, file_name=voice_hat_path, **wav_quality)
+    wav_write(bg_hat, file_name=bg_hat_path, **wav_quality)
 
     return sdr, sir
 
@@ -347,7 +372,7 @@ def _get_data_training(current_set, set_size, mixtures_list, sources_list,
 
 
 def _get_data_testing(sources_parent_path, window_values, fft_size, hop,
-                      seq_length, context_length, batch_size):
+                      seq_length, context_length, batch_size, usage_case):
     """Gets the actual input and output data for testing.
 
     :param sources_parent_path: The parent path of the sources
@@ -364,17 +389,24 @@ def _get_data_testing(sources_parent_path, window_values, fft_size, hop,
     :type context_length: int
     :param batch_size: The batch size.
     :type batch_size: int
+    :param usage_case: Flag to indicate that currently we are just using it.
+    :type usage_case: bool
     :return: The actual input and target value.
     :rtype: numpy.core.multiarray.ndarray
     """
-    bass = wav_read(os.path.join(sources_parent_path, 'bass.wav'), mono=False)[0]
-    drums = wav_read(os.path.join(sources_parent_path, 'drums.wav'), mono=False)[0]
-    others = wav_read(os.path.join(sources_parent_path, 'other.wav'), mono=False)[0]
-    voice = wav_read(os.path.join(sources_parent_path, 'vocals.wav'), mono=False)[0]
-
-    bg_true = np.sum(bass + drums + others, axis=-1) * 0.5
-    mix = np.sum(bass + drums + others + voice, axis=-1) * 0.5
-    voice_true = np.sum(voice, axis=-1) * 0.5
+    if not usage_case:
+        bass = wav_read(os.path.join(sources_parent_path, 'bass.wav'), mono=False)[0]
+        drums = wav_read(os.path.join(sources_parent_path, 'drums.wav'), mono=False)[0]
+        others = wav_read(os.path.join(sources_parent_path, 'other.wav'), mono=False)[0]
+        voice = wav_read(os.path.join(sources_parent_path, 'vocals.wav'), mono=False)[0]
+
+        bg_true = np.sum(bass + drums + others, axis=-1) * 0.5
+        voice_true = np.sum(voice, axis=-1) * 0.5
+        mix = np.sum(bass + drums + others + voice, axis=-1) * 0.5
+    else:
+        mix = wav_read(sources_parent_path, mono=True)[0]
+        voice_true = None
+        bg_true = None
 
     mix_magnitude, mix_phase = stft(mix, window_values, fft_size, hop)
 

helpers/settings.py

@@ -20,9 +20,12 @@
     'testing_output_string_all',
     'training_constants',
     'wav_quality',
-    'hyper_parameters'
+    'hyper_parameters',
+    'usage_output_string_per_example',
+    'usage_output_string_total'
 ]
 
+
 debug = False
 _debug_suffix = '_debug' if debug else ''
 
@@ -62,8 +65,8 @@
 }
 
 metrics_paths = {
-    'sdr': os.path.join(_metrics_path, 'sdr{}.pckl'.format(_debug_suffix)),
-    'sir': os.path.join(_metrics_path, 'sir{}.pckl'.format(_debug_suffix))
+    'sdr': os.path.join(_metrics_path, 'sdr{}_p2.pckl'.format(_debug_suffix)),
+    'sir': os.path.join(_metrics_path, 'sir{}_p2.pckl'.format(_debug_suffix))
 }
 
 output_states_path = {
@@ -87,6 +90,9 @@
                             'Median SIR:{sir:6.2f} dB | ' \
                             'Total time:{t:6.2f} sec(s)'
 
+usage_output_string_per_example = '-- File {f} processed. Time: {t:6.2f} sec(s)'
+usage_output_string_total = '-- All files processed. Total time: {t:6.2f} sec(s)'
+
 # Process constants
 training_constants = {
     'epochs': 2 if debug else 100,

scripts/testing.py

@@ -4,6 +4,8 @@
 """Testing process module.
 """
 
+from __future__ import print_function
+
 import pickle
 import time
 

scripts/training.py

@@ -4,6 +4,8 @@
 """Training process module.
 """
 
+from __future__ import print_function
+
 import time
 
 import torch