Merge branch 'master' into face_tremor

dbm_lib/controller/process_feature.py

@@ -136,7 +136,7 @@ def process_movement(video_uri, out_dir, dbm_group, r_config, dlib_model):
     facial_tremor.fac_tremor_process(video_uri, out_dir, r_config, model_output=True)
 
     
-def process_nlp(video_uri, out_dir, dbm_group, r_config, deep_path):
+def process_nlp(video_uri, out_dir, dbm_group, tran_tog, r_config, deep_path):
     """
     processing nlp features
     Args:
@@ -144,12 +144,12 @@ def process_nlp(video_uri, out_dir, dbm_group, r_config, deep_path):
         dbm_group: list of features to process; r_config: raw feature config object
         deep_path: deep speech build path
     """
-    if dbm_group != None and len(dbm_group)>0 and 'nlp' not in dbm_group:
+    if dbm_group != None and len(dbm_group)>0 and 'speech' not in dbm_group:
         return
     
     logger.info('Processing nlp variables from data in {}'.format(video_uri))
     transcribe.run_transcribe(video_uri, out_dir, r_config, deep_path)
-    speech_features.run_speech_feature(video_uri, out_dir, r_config)
+    speech_features.run_speech_feature(video_uri, out_dir, r_config, tran_tog)
     
 
 def remove_file(file_path):

dbm_lib/dbm_features/raw_features/nlp/speech_features.py

@@ -10,6 +10,7 @@ import pandas as pd
 import glob
 from os.path import join
 import logging
+import shutil
 
 from dbm_lib.dbm_features.raw_features.util import util as ut
 from dbm_lib.dbm_features.raw_features.util import nlp_util as n_util
@@ -21,7 +22,7 @@ speech_dir = 'nlp/speech_feature'
 speech_ext = '_nlp.csv'
 transcribe_ext = 'nlp/transcribe/*_transcribe.csv'
 
-def run_speech_feature(video_uri, out_dir, r_config):
+def run_speech_feature(video_uri, out_dir, r_config, tran_tog):
     """
     Processing all patient's for fetching nlp features
     -------------------
@@ -42,6 +43,9 @@ def run_speech_feature(video_uri, out_dir, r_config):
 
             logger.info('Saving Output file {} '.format(out_loc))
             ut.save_output(df_speech, out_loc, fl_name, speech_dir, speech_ext)
+
+            if (tran_tog == None) or (tran_tog != 'on'):
+                shutil.rmtree(os.path.dirname(transcribe_path[0]))
             
     except Exception as e:
         logger.error('Failed to process video file')

dbm_lib/dbm_features/raw_features/util/nlp_util.py

@@ -126,7 +126,7 @@ def process_speech(transcribe_df,r_config):
         Returns:
             Dataframe for speech features
     """
-    
+    transcribe_df = transcribe_df.replace(np.nan, '', regex=True)
     err_transcribe = transcribe_df[r_config.err_reason].iloc[0]
     transcribe = transcribe_df[r_config.nlp_transcribe].iloc[0]
     total_time = transcribe_df[r_config.nlp_totalTime].iloc[0]

dbm_lib/dbm_features/raw_features/util/vad_utilities.py

@@ -158,7 +158,7 @@ def vad_get_segment_times(sample_rate, frame_duration_ms,
     for frame in frames:
         is_speech = vad.is_speech(frame.bytes, sample_rate)
 
-        sys.stdout.write('1' if is_speech else '0')
+        #sys.stdout.write('1' if is_speech else '0')
         if not triggered:
             ring_buffer.append((frame, is_speech))
             num_voiced = len([f for f, speech in ring_buffer if speech])
@@ -167,7 +167,7 @@ def vad_get_segment_times(sample_rate, frame_duration_ms,
             # TRIGGERED state.
             if num_voiced > 0.9 * ring_buffer.maxlen:
                 triggered = True
-                sys.stdout.write('+(%s)' % (ring_buffer[0][0].timestamp,))
+                #sys.stdout.write('+(%s)' % (ring_buffer[0][0].timestamp,))
                 start_times.append(ring_buffer[0][0].timestamp)  # BT
                 ring_buffer.clear()
         else:
@@ -179,18 +179,18 @@ def vad_get_segment_times(sample_rate, frame_duration_ms,
             # unvoiced, then enter NOTTRIGGERED and yield whatever
             # audio we've collected.
             if num_unvoiced > 0.9 * ring_buffer.maxlen:
-                sys.stdout.write('-(%s)' % (frame.timestamp + frame.duration))
+                #sys.stdout.write('-(%s)' % (frame.timestamp + frame.duration))
                 end_times.append(ring_buffer[0][0].timestamp + frame.duration)  # BT
                 triggered = False
 
     if triggered:  # BT if were in triggered state at end of signal, set output time
-        sys.stdout.write('-(%s)' % (frame.timestamp + frame.duration))
+        #sys.stdout.write('-(%s)' % (frame.timestamp + frame.duration))
         if len(ring_buffer)>0:
             end_times.append(ring_buffer[0][0].timestamp )  # BT
         else:
             # only get here in very rare case that we triggered on 2nd-to-last frame
             end_times.append(frame.timestamp + frame.duration)
-    sys.stdout.write('\n')
+    #sys.stdout.write('\n')
 
     return(start_times, end_times)