blackteay/ffmpeg-python
1
0
Fork 0
mirror of https://github.com/kkroening/ffmpeg-python.git synced 2025-04-06 04:15:44 +08:00
Code Issues Projects Releases Wiki Activity

Add --max-silence-time in split_silence.py

Browse Source
This commit is contained in:
Karl Kroening 2018-01-16 00:18:57 -08:00
parent dad4c3cb55
commit 4bb2fc4999
1 changed files with 9 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
examples/split_silence.py
View File

@ -29,6 +29,7 @@ parser.add_argument('--start-time', type=float, help='Start time (seconds)')
parser.add_argument('--end-time', type=float, help='End time (seconds)') parser.add_argument('--end-time', type=float, help='End time (seconds)')
parser.add_argument('--padding', type=float, default=0., help='Output silence padding (seconds)') parser.add_argument('--padding', type=float, default=0., help='Output silence padding (seconds)')
parser.add_argument('--min-chunk-time', type=float, default=0., help='Minimum chunk duration') parser.add_argument('--min-chunk-time', type=float, default=0., help='Minimum chunk duration')
parser.add_argument('--max-silence-time', type=float, help='Maximum silence time before guaranteed chunk splitting')
parser.add_argument('--metadata-filename', help='Optional metadata output file') parser.add_argument('--metadata-filename', help='Optional metadata output file')
parser.add_argument('-v', dest='verbose', action='store_true', help='Verbose mode') parser.add_argument('-v', dest='verbose', action='store_true', help='Verbose mode')
@ -110,7 +111,7 @@ def _makedirs(path):
raise raise
def combine_chunks(chunk_times, min_chunk_time): def combine_chunks(chunk_times, min_chunk_time, max_silence_time=None):
chunk_times = copy.copy(chunk_times) chunk_times = copy.copy(chunk_times)
pop_chunk = lambda chunk_times: (chunk_times[0][0], chunk_times[0][1], chunk_times[1:]) pop_chunk = lambda chunk_times: (chunk_times[0][0], chunk_times[0][1], chunk_times[1:])
new_chunk_times = [] new_chunk_times = []
@ -120,7 +121,11 @@ def combine_chunks(chunk_times, min_chunk_time):
time = end_time - start_time time = end_time - start_time
while time < min_chunk_time and len(chunk_times) != 0: while time < min_chunk_time and len(chunk_times) != 0:
# Combine with next chunk. # Combine with next chunk.
_, end_time, chunk_times = pop_chunk(chunk_times) next_start_time, next_end_time, chunk_times = pop_chunk(chunk_times)
if max_silence_time is not None and next_start_time - end_time > max_silence_time:
new_chunk_times.append((start_time, end_time))
start_time = next_start_time
end_time = next_end_time
time = end_time - start_time time = end_time - start_time
new_chunk_times.append((start_time, end_time)) new_chunk_times.append((start_time, end_time))
return new_chunk_times return new_chunk_times
@ -135,12 +140,13 @@ def split_audio(
end_time=None, end_time=None,
padding=0., padding=0.,
min_chunk_time=0., min_chunk_time=0.,
max_silence_time=None,
metadata_filename=None, metadata_filename=None,
verbose=False, verbose=False,
): ):
chunk_times = get_chunk_times(in_filename, silence_threshold, silence_duration, start_time, end_time) chunk_times = get_chunk_times(in_filename, silence_threshold, silence_duration, start_time, end_time)
if min_chunk_time > 0.: if min_chunk_time > 0.:
chunk_times = combine_chunks(chunk_times, min_chunk_time) chunk_times = combine_chunks(chunk_times, min_chunk_time, max_silence_time)
metadata = [] metadata = []
for i, (start_time, end_time) in enumerate(chunk_times): for i, (start_time, end_time) in enumerate(chunk_times):