def get_fingerprint(filename):
audio_part = AudioSegment.from_file(filename, "mp4")
result_fingerprint = work.work_audio(audio_part, filename)
result = result_fingerprint[0]
frame_rate = result_fingerprint[1]
resultant = list(result)
return [resultant, frame_rate]
else:
i = i + 1
result_previous = None
continue
else:
filename1 = list_videos[i - 1].video_clip
filename2 = list_videos[i].video_clip
audio1 = AudioSegment.from_file(filename1, "mp4")
audio2 = AudioSegment.from_file(filename2, "mp4")
if result_previous is not None:
result1_1 = result_previous
else:
result1_1 = work.work_audio(audio1, filename1)
result2_2 = work.work_audio(audio2, filename2)
result1 = result1_1[0]
result2 = result2_2[0]
frame_rate = result1_1[1]
result_previous = result2_2
res2 = list(result2)
res2.sort(key=lambda x: x[1])
res1 = list(result1)
l = {}
print("Length of result 1: " + str(len(res1)))
print("Length of result 2: " + str(len(res2)))
# for tup in res2:
def audio_main(video_list):
file_file = open(sys.argv[1] + "/mashup/audio_names.txt", "r")
line = file_file.readline()
list_videos = []
while line:
line = line.rstrip()
list_videos.append(line)
line = file_file.readline()
#list_videos = ["video1.mp4", "video2.mp4"]
print(list_videos)
list_time = []
#audiofile1 = AudioSegment.from_mp3(filename)
i = 1
result_previous = None
if (list_videos.__len__() == 1):
audio_file = AudioSegment.from_file(list_videos[0], "mp4")
final_audio = audio_file
final_audio.export(sys.argv[1] + "/mashup/mashup.mp3", format="mp3")
return
starting = 1
while i < list_videos.__len__():
filename1 = list_videos[i - 1]
filename2 = list_videos[i]
audio1 = AudioSegment.from_file(filename1, "mp4")
audio2 = AudioSegment.from_file(filename2, "mp4")
if result_previous is not None:
result1_1 = result_previous
else:
result1_1 = work.work_audio(audio1, filename1)
result2_2 = work.work_audio(audio2, filename2)
result1 = result1_1[0]
result2 = result2_2[0]
frame_rate = result1_1[1]
result_previous = result2_2
res2 = list(result2)
res2.sort(key=lambda x: x[1])
res1 = list(result1)
l = {}
for tup in res2:
for tu in res1:
if tu[0] == tup[0]:
if (tu[1] - tup[1]) in l:
l[tu[1] - tup[1]] = l[tu[1] - tup[1]] + 1
else:
l[tu[1] - tup[1]] = 1
maxi = 0
max_index = 0
for key in l:
if l[key] > maxi:
maxi = l[key]
max_index = key
print(maxi)
print(max_index)
match_value = (2048 * max_index * 1000) / frame_rate
match_sec = match_value / 1000.0
if match_value < 0:
list_videos[i -
1], list_videos[i] = list_videos[i], list_videos[i - 1]
#video_list[i-1], video_list[i] = video_list[i], video_list[i-1]
"""
result_previous = result1_1
if starting == 1:
video_list[i-1].start_time = 0
video_list[i-1].end_time = audio2.duration_seconds
starting = 0
video_list[i].start_time = abs(match_sec) + video_list[i - 1].start_time
video_list[i].end_time = video_list[i].start_time + audio1.duration_seconds
else:
video_list[i].start_time = abs(match_sec) + video_list[i-1].start_time
video_list[i].end_time = video_list[i].start_time + audio2.duration_seconds
"""
print(match_value)
list_time.append(abs(match_value))
i = i + 1
print(list_videos)
#Normalizing audio
max_vol = 0
for i in range(0, len(list_videos)):
audio = AudioSegment.from_file(list_videos[i], "mp4")
if max_vol == 0:
max_vol = audio.dBFS
if audio.dBFS > max_vol:
max_vol = audio.dBFS
#Normalizing audio
final_audio_temp = AudioSegment.from_file(list_videos[0], "mp4")
final_audio_temp = final_audio_temp.apply_gain(max_vol -
final_audio_temp.dBFS)
final_audio = final_audio_temp[:(list_time[0] + 1000)]
i = 1
while i < list_time.__len__():
audio_file = AudioSegment.from_file(list_videos[i], "mp4")
audio_file = audio_file.apply_gain(max_vol - audio_file.dBFS)
#final_audio = final_audio + audio_file[:list_time[i]]
final_audio = final_audio.append(audio_file[:(list_time[i] + 1000)],
crossfade=1000)
i = i + 1
audio_file = AudioSegment.from_file(list_videos[i], "mp4")
audio_file = audio_file.apply_gain(max_vol - audio_file.dBFS)
#final_audio = final_audio + audio_file
final_audio = final_audio.append(audio_file, crossfade=1000)
final_audio.export(sys.argv[1] + "/mashup/mashup.mp3", format="mp3")
return video_list
def audio_main():
file_file = open("audio_names.txt", "r")
line = file_file.readline()
list_videos = []
while line:
line = line.rstrip()
list_videos.append(line)
line = file_file.readline()
#list_videos = ["video1.mp4", "video2.mp4"]
print(list_videos)
list_time = []
audiofile1 = AudioSegment.from_mp3(filename)
i = 1
result_previous = None
if (list_videos.__len__() == 1):
audio_file = AudioSegment.from_file(list_videos[0], "mp4")
final_audio = audio_file
final_audio.export("mashup.mp3", format="mp3")
return
while i < list_videos.__len__():
filename1 = list_videos[i - 1]
filename2 = list_videos[i]
audio1 = AudioSegment.from_file(filename1, "mp4")
audio2 = AudioSegment.from_file(filename2, "mp4")
if result_previous is not None:
result1_1 = result_previous
else:
result1_1 = work.work_audio(audio1, filename1)
result2_2 = work.work_audio(audio2, filename2)
result1 = result1_1[0]
result2 = result2_2[0]
frame_rate = result1_1[1]
result_previous = result2_2
res2 = list(result2)
res2.sort(key=lambda x: x[1])
res1 = list(result1)
l = {}
for tup in res2:
for tu in res1:
if tu[0] == tup[0]:
if (tu[1] - tup[1]) in l:
l[tu[1] - tup[1]] = l[tu[1] - tup[1]] + 1
else:
l[tu[1] - tup[1]] = 1
maxi = 0
max_index = 0
for key in l:
if l[key] > maxi:
maxi = l[key]
max_index = key
print(maxi)
print(max_index)
match_value = (2048 * max_index * 1000) / frame_rate
if match_value < 0:
list_videos[i -
1], list_videos[i] = list_videos[i], list_videos[i - 1]
print(match_value)
list_time.append(abs(match_value))
i = i + 1
print(list_videos)
i = 1
final_audio_temp = AudioSegment.from_file(list_videos[0], "mp4")
final_audio = final_audio_temp[:list_time[0]]
while i < list_time.__len__():
audio_file = AudioSegment.from_file(list_videos[i], "mp4")
final_audio = final_audio + audio_file[:list_time[i]]
i = i + 1
audio_file = AudioSegment.from_file(list_videos[i], "mp4")
final_audio = final_audio + audio_file
final_audio.export("mashup.mp3", format="mp3")
return