Update app.py

app.py

@@ -6,7 +6,7 @@ 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):
-    audio = AudioSegment.from_file(audio_fp).set_channels(1)
+    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))
     speech_timestamps = [{"start": s[0]/1000, "end": s[1]/1000} for s in speech_timestamps]
@@ -66,7 +66,7 @@ def get_labels(audio_fp, min_speech_duration_ms, min_silence_duration_ms, auto_m
 interface = gr.Interface(
     get_labels,
     [
-        gr.Audio(type="filepath", label="Audio file"),
+        gr.File(type="filepath", label="Audio file", file_types=["audio"], file_count="single"),
         gr.Number(label="min_speech_duration_ms", value=40, info="default (40)"), 
         gr.Number(label="min_silence_duration_ms", value=40, info="default (40)"),
         gr.Checkbox(label="Auto merge", value=True),