Update app.py

app.py

@@ -5,7 +5,7 @@ from pydub.silence import detect_nonsilent
 from pydub import AudioSegment
  
 
-def get_labels(audio_fp, min_speech_duration_ms, min_silence_duration_ms, auto_merge, uppper_merge_threshold, max_segment_length, end_extension, extend_small_segments):
+def get_labels(audio_fp, min_speech_duration_ms, min_silence_duration_ms, auto_merge, uppper_merge_threshold, max_segment_length, end_extension, extend_small_segments, show_duration_label):
     audio = AudioSegment.from_file(audio_fp.name).set_channels(1)
     speech_timestamps = detect_nonsilent(audio, min_silence_len=min_silence_duration_ms, silence_thresh=-40)
     speech_timestamps = list(filter(lambda x: x[1]-x[0] > min_speech_duration_ms, speech_timestamps))
@@ -57,6 +57,10 @@ def get_labels(audio_fp, min_speech_duration_ms, min_silence_duration_ms, auto_m
                 if nl[1] + end_extension < new_labels[i+1][0]:
                     nl[1] = nl[1] + end_extension
 
+    if show_duration_label:
+        for nl in new_labels:
+            nl[2] = round(nl[1] - nl[0], 3)
+
     translate_labels = list(map(lambda x: f"{x[0]}\t{x[1]}\t{x[2]}", new_labels))
 
     filename_path = f"{fn}_translate_label.txt"
@@ -76,7 +80,8 @@ interface = gr.Interface(
         gr.Textbox(label="Gap max threshold value (seconds)", value=0.350),
         gr.Number(label="Approx Max Segment Length", value=7),
         gr.Number(label="Extend end by (seconds)", value=0),
-        gr.Checkbox(label="Extend small segments (minimum 1.02 seconds)", value=False)
+        gr.Checkbox(label="Extend small segments (minimum 1.02 seconds)", value=False),
+        gr.Checkbox(label="Show only duration in labels", value=False)
     ],
     [
         gr.File(label="VAD Labels"),