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 --padding in split_silence.py

Browse Source
This commit is contained in:
Karl Kroening 2018-01-14 23:13:54 -08:00
parent 87f85000ac
commit e5003648bb
1 changed files with 21 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

32
examples/split_silence.py
View File

@ -26,8 +26,9 @@ parser.add_argument('--silence-threshold', default=DEFAULT_THRESHOLD, type=int,
parser.add_argument('--silence-duration', default=DEFAULT_DURATION, type=float, help='Silence duration') parser.add_argument('--silence-duration', default=DEFAULT_DURATION, type=float, help='Silence duration')
parser.add_argument('--start-time', type=float, help='Start time (seconds)') 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('-v', dest='verbose', action='store_true', help='Verbose mode') parser.add_argument('--padding', type=float, default=0., help='Output silence padding (seconds)')
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')
silence_start_re = re.compile(' silence_start: (?P<start>[0-9]+(\.?[0-9]*))$') silence_start_re = re.compile(' silence_start: (?P<start>[0-9]+(\.?[0-9]*))$')
@ -50,7 +51,7 @@ def get_chunk_times(in_filename, silence_threshold, silence_duration, start_time
if end_time is not None: if end_time is not None:
input_kwargs['t'] = end_time - start_time input_kwargs['t'] = end_time - start_time
p = _logged_popen( child = _logged_popen(
(ffmpeg (ffmpeg
.input(in_filename, **input_kwargs) .input(in_filename, **input_kwargs)
.filter_('silencedetect', n='{}dB'.format(silence_threshold), d=silence_duration) .filter_('silencedetect', n='{}dB'.format(silence_threshold), d=silence_duration)
@ -59,8 +60,8 @@ def get_chunk_times(in_filename, silence_threshold, silence_duration, start_time
) + ['-nostats'], # FIXME: use .nostats() once it's implemented in ffmpeg-python. ) + ['-nostats'], # FIXME: use .nostats() once it's implemented in ffmpeg-python.
stderr=subprocess.PIPE stderr=subprocess.PIPE
) )
output = p.communicate()[1].decode('utf-8') output = child.communicate()[1].decode('utf-8')
if p.returncode != 0: if child.returncode != 0:
sys.stderr.write(output) sys.stderr.write(output)
sys.exit(1) sys.exit(1)
logger.debug(output) logger.debug(output)
@ -114,6 +115,7 @@ def split_audio(
silence_duration=DEFAULT_DURATION, silence_duration=DEFAULT_DURATION,
start_time=None, start_time=None,
end_time=None, end_time=None,
padding=0.,
metadata_filename=None, metadata_filename=None,
verbose=False, verbose=False,
): ):
@ -125,9 +127,9 @@ def split_audio(
out_filename = out_pattern.format(i, i=i) out_filename = out_pattern.format(i, i=i)
_makedirs(os.path.dirname(out_filename)) _makedirs(os.path.dirname(out_filename))
start_text = '{:.02f}'.format(start_time) start_text = '{:.04f}'.format(start_time)
end_text = '{:.02f}'.format(end_time) end_text = '{:.04f}'.format(end_time)
duration_text = '{:.02f}'.format(time) duration_text = '{:.04f}'.format(time)
metadata.append({ metadata.append({
'filename': out_filename, 'filename': out_filename,
'start': start_text, 'start': start_text,
@ -136,16 +138,24 @@ def split_audio(
}) })
logger.info('{}: start={}, end={}, duration={}'.format(out_filename, start_text, end_text, duration_text)) logger.info('{}: start={}, end={}, duration={}'.format(out_filename, start_text, end_text, duration_text))
_logged_popen( input = ffmpeg.input(in_filename, ss=start_time, t=time)
(ffmpeg if padding > 0.:
.input(in_filename, ss=start_time, t=time) silence = ffmpeg.input('anullsrc', format='lavfi', t=padding)
input = ffmpeg.concat(silence, input, silence, v=0, a=1)
child = _logged_popen(
(input
.output(out_filename) .output(out_filename)
.overwrite_output() .overwrite_output()
.compile() .compile()
), ),
stdout=subprocess.PIPE if not verbose else None, stdout=subprocess.PIPE if not verbose else None,
stderr=subprocess.PIPE if not verbose else None, stderr=subprocess.PIPE if not verbose else None,
).communicate() )
out = child.communicate()
if child.returncode != 0:
if not verbose:
sys.stderr.write(out[1].decode('utf-8'))
if metadata_filename is not None: if metadata_filename is not None:
_makedirs(os.path.dirname(metadata_filename)) _makedirs(os.path.dirname(metadata_filename))