I would like to run the following command in a python script, I also want to make it loop over several videos in a folder. This is the command I want to run.
ffmpeg -i mymovie.avi -f image2 -vf fps=fps=1 output%d.png
I want to fit it in something like this:
import ffmpy
import os
path = './Videos/MyVideos/'
for filename in os.listdir(path):
name = filename.replace('.avi','')
os.mkdir(os.path.join(path,name))
*ffmpeg command here*
I found a wrapper for ffmpeg called ffmpy, could this be a solution?
From a brief look at FFMPY, you could do this using ffmpy.FFmpeg, as that allows any and all FFMPEG command line options, including -f. -- Click the link for documentation.
You could do the FFMPEG command with os.system. You'll need to import OS anyway to iterate through the files.
You would need to iterate through all the files in a directory though. This would be the more challenging bit, it's quite easy with a for loop though.
for filename in os.listdir(path):
if (filename.endswith(".mp4")): #or .avi, .mpeg, whatever.
os.system("ffmpeg -i {0} -f image2 -vf fps=fps=1 output%d.png".format(filename))
else:
continue
The above code iterates through the directory at path, and uses command prompt to execute your given FFMPEG command, using the filename (if it's a video file) in place of mymovie.avi
Dont have reputation to comment, hence adding another response.
Another version of ocelot's answer with the more readable f-string syntax of python -
for filename in os.listdir(path):
if (filename.endswith(".mp4")): #or .avi, .mpeg, whatever.
os.system(f'ffmpeg -i {filename} -f image2 -vf fps=fps=1 output%d.png')
else:
continue
Try pydemux in https://github.com/Tee0125/pydemux. Pydemux module can extract video frames as in Pillow Image format
from PyDemux import Video
v = Video.open('video.mov')
i = 0
while True:
im = v.get_frame()
if im is None:
break
im.save('output%d.png'%i)
i = i + 1
This is a way to use ffmpeg in a python script, e.g. for extracting the last 10 seconds of every video:
ffmpeg -sseof -10 -i input.mp4 output.mp4
To apply this to a whole folder of mp4 files:
from pathlib import Path
import os
suffix = ".mp4"
input_path= Path.home() / "Desktop/foo"
file_paths= [subp for subp in input_path.rglob('*') if suffix == subp.suffix]
file_paths.sort()
output_path = Path.home() / "Desktop/foo/new"
output_path.mkdir(parents=True, exist_ok=True)
for file_p in file_paths:
input = str(file_p)
output = str( output_path / file_p.name )
command = f"ffmpeg -sseof -10 -i {input} {output}"
print(command)
os.system(command)
Related
i've been trying to run sox in a python script but it can't the output file and gives me [errno2]
def AudioToSpectrogram(self, files, pixel_per_sec, height, width):
file_name = ("tmp_{}.png").format(random.randint(0, 100000))
command = "sox -V0 {} -n remix 1 rate 10k spectrogram -y {} -x {} -X {} -m -r -o {}".format(files, height, width, pixel_per_sex, file_name)
p = Popen(command, shell=True, stdin=PIPE, stdout=PIPE, stderr=STDOUT, close_fds=True)
Output, errors = p.communicate()
If errors:
Print(errors)
Image = Image.open(file_name)
Os.remove(file_name)
Return np.array(image)
This is the error it gives
Exception: [errno2] No such file or Directory: 'tmp_47483.png'
I hope you could give me some pointers since i am still new in this field, thanks in advance!
Assuming tmp_47483.png is in fact being created, the problem is likely that the command is placing the file in a different folder and Python can't find it inside the current working directory. The
# manually set the full path (make sure the backslashes are escaped by putting two each)
file_name = f"C:\\Full\\Path\\To\\File\\tmp_{random.randint(0, 100000}.png"
# use the os module to join the path
base_dir = "C:\\Full\\Path\\To\\File"
file_name = os.path.join(base_dir, f"tmp_{random.randint(0, 100000}.png")
# if you want it to appear in the same folder as your script:
CWD = os.path.dirname(os.path.realpath(__file__)) # mostly fool-proof way of getting a script's Current Working Directory
file_name = os.path.join(CWD, f"tmp_{random.randint(0, 100000}.png")
Try these and see if they help. If not, make sure that command is actually working and outputting a file somewhere.
I'm trying to combine a video(with no sound) and its separate audio file
I've tried ffmpeg ffmpeg -i video.mp4 -i audio.mp4 -c copy output.mp4
and it works fine.
i'm trying to achieve the same output from ffmpeg-python but with no luck. Any help on how to do this?
I had the same problem.
Here is the python code after you have pip install ffmpeg-python in your environment:
import ffmpeg
input_video = ffmpeg.input('./test/test_video.webm')
input_audio = ffmpeg.input('./test/test_audio.webm')
ffmpeg.concat(input_video, input_audio, v=1, a=1).output('./processed_folder/finished_video.mp4').run()
v=1:
Set the number of output video streams, that is also the number of video streams in each segment. Default is 1.
a=1: Set the number of output audio streams, that is also the number of audio streams in each segment. Default is 0.
For the details of ffmpeg.concat, check out: https://ffmpeg.org/ffmpeg-filters.html#concat.
You can check out more examples here: https://github.com/kkroening/ffmpeg-python/issues/281
Hope this helps!
PS.
If you are using MacOS and have the error:
FileNotFoundError: [Errno 2] No such file or directory: 'ffmpeg' while running the code, just brew install ffmpeg in your terminal.
You could use subprocess:
import subprocess
subprocess.run("ffmpeg -i video.mp4 -i audio.mp4 -c copy output.mp4")
You can also use fstrings to use variable names as input:
videofile = "video.mp4"
audiofile = "audio.mp4"
outputfile = "output.mp4"
codec = "copy"
subprocess.run(f"ffmpeg -i {videofile} -i {audiofile} -c {codec} {outputfile}")
import ffmpeg
input_video = ffmpeg.input("../resources/video_with_audio.mp4")
added_audio = ffmpeg.input("../resources/dance_beat.ogg").audio.filter('adelay', "1500|1500")
merged_audio = ffmpeg.filter([input_video.audio, added_audio], 'amix')
(ffmpeg
.concat(input_video, merged_audio, v=1, a=1)
.output("mix_delayed_audio.mp4")
.run(overwrite_output=True))
you can review this link https://github.com/kkroening/ffmpeg-python/issues/281#issuecomment-546724993
Added the following code:
https://github.com/russellstrei/combineViaFFMPEG
It walks the directory, finds ".mp4" files, adds it to the file to be used by ffmpeg, then executes the command.
for name in glob.glob(directory +"\\"+ '*.mp4'):
print(name)
file1 = open(processList, "a") # append mode
file1.write("file '" + name + "'\n")
file1.close()
execute()
def execute():
cmd = "ffmpeg -f concat -safe 0 -i " + processList + " -c copy "+ dir + "output.mp4"
os.system(cmd)
For instance, i have five .mp4 in a folder. How do i loop and convert them to .mp3 and output them in a different folder?
Thanks
I have tried this but it is not working.
**import moviepy.editor as mp
import re
tgt_folder = "D:/Music/video"
for file in [n for n in os.listdir(tgt_folder) if re.search('mp4',n)]:
full_path = os.path.join(tgt_folder, file)
output_path = os.path.join(tgt_folder, os.path.splitext(file)[0] + '.mp3')
clip = mp.AudioFileClip(full_path).subclip(10,) # disable if do not want any clipping
clip.write_audiofile('D:/Music/mp3')**
you can do this by using ffmpeg and python subprocces
import subprocess
subprocess.run(["ffmpeg", "-i video.mp4 -b:a 192K -vn music.mp3"])
I'm trying to combine a video(with no sound) and its separate audio file
I've tried ffmpeg ffmpeg -i video.mp4 -i audio.mp4 -c copy output.mp4
and it works fine.
i'm trying to achieve the same output from ffmpeg-python but with no luck. Any help on how to do this?
I had the same problem.
Here is the python code after you have pip install ffmpeg-python in your environment:
import ffmpeg
input_video = ffmpeg.input('./test/test_video.webm')
input_audio = ffmpeg.input('./test/test_audio.webm')
ffmpeg.concat(input_video, input_audio, v=1, a=1).output('./processed_folder/finished_video.mp4').run()
v=1:
Set the number of output video streams, that is also the number of video streams in each segment. Default is 1.
a=1: Set the number of output audio streams, that is also the number of audio streams in each segment. Default is 0.
For the details of ffmpeg.concat, check out: https://ffmpeg.org/ffmpeg-filters.html#concat.
You can check out more examples here: https://github.com/kkroening/ffmpeg-python/issues/281
Hope this helps!
PS.
If you are using MacOS and have the error:
FileNotFoundError: [Errno 2] No such file or directory: 'ffmpeg' while running the code, just brew install ffmpeg in your terminal.
You could use subprocess:
import subprocess
subprocess.run("ffmpeg -i video.mp4 -i audio.mp4 -c copy output.mp4")
You can also use fstrings to use variable names as input:
videofile = "video.mp4"
audiofile = "audio.mp4"
outputfile = "output.mp4"
codec = "copy"
subprocess.run(f"ffmpeg -i {videofile} -i {audiofile} -c {codec} {outputfile}")
import ffmpeg
input_video = ffmpeg.input("../resources/video_with_audio.mp4")
added_audio = ffmpeg.input("../resources/dance_beat.ogg").audio.filter('adelay', "1500|1500")
merged_audio = ffmpeg.filter([input_video.audio, added_audio], 'amix')
(ffmpeg
.concat(input_video, merged_audio, v=1, a=1)
.output("mix_delayed_audio.mp4")
.run(overwrite_output=True))
you can review this link https://github.com/kkroening/ffmpeg-python/issues/281#issuecomment-546724993
Added the following code:
https://github.com/russellstrei/combineViaFFMPEG
It walks the directory, finds ".mp4" files, adds it to the file to be used by ffmpeg, then executes the command.
for name in glob.glob(directory +"\\"+ '*.mp4'):
print(name)
file1 = open(processList, "a") # append mode
file1.write("file '" + name + "'\n")
file1.close()
execute()
def execute():
cmd = "ffmpeg -f concat -safe 0 -i " + processList + " -c copy "+ dir + "output.mp4"
os.system(cmd)
I am trying to loop over a directory of sub folders where every folder contains one .avi file that i want to retrieve its length in seconds.
I've found PyMedia http://pymedia.org/ and i understand it could possibly help me achieve this but i cannot find anything about avi duration / length in the documentation.
How would i be able to do that? also, if there is a different library of some sort i'd like to know aswel.
Edit: Added my final solution that works thanks to J.F. Sebastian
import sys
import glob
import os
from hachoir_core.cmd_line import unicodeFilename
from hachoir_core.i18n import getTerminalCharset
from hachoir_metadata import extractMetadata
from hachoir_parser import createParser
path = "z:\*"
for fpath in glob.glob(os.path.join(path, '*avi')):
filename = fpath
filename, real_filename = unicodeFilename(filename), filename
parser = createParser(filename, real_filename=real_filename)
metadata = extractMetadata(parser)
print fpath
print("Duration (hh:mm:ss.f): %s" % metadata.get('duration'))
print '\n'
You could use hachoir-metadata to extract avi duration from a file:
#!/usr/bin/env python
import sys
# $ pip install hachoir-{core,parser,metadata}
from hachoir_core.cmd_line import unicodeFilename
from hachoir_core.i18n import getTerminalCharset
from hachoir_metadata import extractMetadata
from hachoir_parser import createParser
filename = sys.argv[1]
charset = getTerminalCharset()
filename, real_filename = unicodeFilename(filename, charset), filename
parser = createParser(filename, real_filename=real_filename)
metadata = extractMetadata(parser)
print("Duration (hh:mm:ss.f): %s" % metadata.get('duration'))
It uses pure Python RIFF parser to extract info from avi file.
Example:
$ get-avi-duration.py test.avi
Duration (hh:mm:ss.f): 0:47:03.360000
Here's ffmpeg's output for comparison:
$ ffmpeg -i test.avi |& grep -i duration
Duration: 00:47:03.36, start: 0.000000, bitrate: 1038 kb/s
To print info about all avi files in a directory tree:
#!/usr/bin/env python
import os
import sys
from hachoir_metadata import extractMetadata
from hachoir_parser import createParser
def getinfo(rootdir, extensions=(".avi", ".mp4")):
if not isinstance(rootdir, unicode):
rootdir = rootdir.decode(sys.getfilesystemencoding())
for dirpath, dirs, files in os.walk(rootdir):
dirs.sort() # traverse directories in sorted order
files.sort()
for filename in files:
if filename.endswith(extensions):
path = os.path.join(dirpath, filename)
yield path, extractMetadata(createParser(path))
for path, metadata in getinfo(u"z:\\"):
if metadata.has('duration'):
print(path)
print(" Duration (hh:mm:ss.f): %s" % metadata.get('duration'))
If your server running any UNIX operation system you can use ffmpeg to do this. Usually just default command like ffmpeg myvideo.avi will give you full video details.
There's also a python wrapper for ffmpeg which probably will return video details in dictionary or list.
EDIT:
I've also found nice ffmpeg tool called ffprobe which can output length of video without additional fuss.
fprobe -loglevel error -show_streams inputFile.avi | grep duration | cut -f2 -d=
Not sure if there is a platform independent way to do this, but if you only need this to work on windows then it looks like MediaInfo (below) has a command line interface which you can use to output details about video files, which could then be parsed to get the information. Not the prettiest solution but looks like it should work.
http://mediainfo.sourceforge.net/en