#coding: utf-8
import os
import cv2
import sys
import json
import argparse
import subprocess
import numpy as np
import moviepy.editor as mp
from PIL import Image
from ..utils._colorings import toBLUE, toGREEN, toACCENT
from ..utils.argparse_utils import ListParamProcessorCreate
from ..utils.generic_utils import now_str
from ..utils.monitor_utils import ProgressMonitor
from ..utils.pil_utils import draw_text_in_pil
[docs]def lyricVideo(argv=sys.argv[1:]):
"""Create a lyric Video.
Args:
json (str) : Path to parameter json file.
--ttfontname (str) : A filename or file-like object containing a TrueType font. If the file is not found in this filename, the loader may also search in other directories, such as the ``fonts/`` directory on Windows or ``/Library/Fonts/`` , ``/System/Library/Fonts/`` and ``~/Library/Fonts/`` on macOS.
--margin (int) : The margin size.
--mode (str) : Optional mode to use for color values.
--fontsize (int) : The font size.
--fontwidth (int) : The font width.
--fontheight (int) : The font height.
--img-size (tuple) : The image size.
--bgRGB (tuple) : he color of background image. (RGB)
--textRGB (tuple) : The color of text. (RGB)
--alpha-range (float) : How many seconds to set alpha to 1 (maximum).
--fps (float) : The video fps.
--span (int) : The span between lyrics.
.. code-block:: python
>>> # Create a json.
>>> import numpy as np
>>> def func(word, start, end, indent='\\t'*4):
... print(f'{indent}"words": "{word}",')
... print(f'{indent}"seconds": {[round(e,3) for e in np.linspace(start, end, len(word))]},')
.. code-block:: python
>>> import json
>>> from pycharmers.utils import dumps_json
>>> with open("era-it-doo/era-it.json") as f:
... data = json.load(f)
>>> print(dumps_json(data))
{
"kwargs": {
"ttfontname": "/Users/iwasakishuto/Library/Fonts/851MkPOP_002.ttf",
"img_size": [
360,
640
],
"fontsize": 38,
"fontwidth": 30,
"margin": 10,
"bgRGB": [
0,
0,
0,
255
],
"mode": "RGBA",
"ret_position": "line"
},
"texts": [
[{"words":"１日今日も、","seconds":[2.4, 2.624, 2.848, 3.072, 3.296, 3.52],"x":-1,"y":50},{"words":"こなした労働、","seconds":[4.1, 4.317, 4.533, 4.75, 4.967, 5.183, 5.4],"x":-1,"y":-1},{"words":"辛かったけど","seconds":[5.7, 5.96, 6.22, 6.48, 6.74, 7.0],"x":-1,"y":-1},{"words":"本当がんばったじゃん","seconds":[7.0, 7.222, 7.444, 7.667, 7.889, 8.111, 8.333, 8.556, 8.778, 9.0],"x":-1,"y":-1},{"words":"１日今日も、","seconds":[9.1, 9.38, 9.66, 9.94, 10.22, 10.5],"x":-1,"y":260},{"words":"いちいちどうこう、","seconds":[10.8, 10.962, 11.125, 11.288, 11.45, 11.612, 11.775, 11.938, 12.1],"x":-1,"y":-1},{"words":"いわれたけど","seconds":[12.3, 12.52, 12.74, 12.96, 13.18, 13.4],"x":-1,"y":-1},{"words":"本当にがんばったじゃん","seconds":[13.4, 13.59, 13.78, 13.97, 14.16, 14.35, 14.54, 14.73, 14.92, 15.11, 15.3],"x":-1,"y":-1}]
]
}
Note:
When you run from the command line, execute as follows::
$ lyricVideo dodo-era-it.json --audio dodo-era-it.mp4[.mp3]
+----------------------------------------+
| Sample |
+========================================+
| .. image:: _images/cli.lyricVideo.gif |
+----------------------------------------+
"""
parser = argparse.ArgumentParser(prog="lyricVideo", description="Create a lyric Video.", add_help=True)
parser.add_argument("json", type=str, help="Path to parameter json file.")
parser.add_argument("--ttfontname", type=str, default=None, help="A filename or file-like object containing a TrueType font. If the file is not found in this filename, the loader may also search in other directories, such as the ``fonts/`` directory on Windows or ``/Library/Fonts/`` , ``/System/Library/Fonts/`` and ``~/Library/Fonts/`` on macOS.")
parser.add_argument("--margin", type=int, default=None, help="The margin size.")
parser.add_argument("--mode", type=str, default=None, help="Optional mode to use for color values.")
parser.add_argument("--fontsize", type=int, default=None, help="The font size.")
parser.add_argument("--fontwidth", type=int, default=None, help="The font width.")
parser.add_argument("--fontheight", type=int, default=None, help="The font height.")
parser.add_argument("--img-size", action=ListParamProcessorCreate(type=int), default=None, help="The image size.")
parser.add_argument("--bgRGB", action=ListParamProcessorCreate(type=int), default=None, help="The color of background image. (RGB)")
parser.add_argument("--textRGB", action=ListParamProcessorCreate(type=int), default=None, help="The color of text. (RGB)")
parser.add_argument("--alpha-range", type=float, help="The video length [s].", default=1.)
parser.add_argument("--fps", type=float, help="The video fps.", default=30.)
parser.add_argument("--span", type=int, help="The span between lyrics", default=None)
parser.add_argument("--audio", type=str, help="The audio path.", default=None)
args = parser.parse_args()
json_path = args.json
with open(json_path, mode="r") as f:
data = json.load(f)
args_kwargs = dict(args._get_kwargs())
data_kwargs = data.get("kwargs", {})
get_kwargs = lambda x,default=None:args_kwargs.get(x) or data_kwargs.get(x, default)
img_size = tuple(get_kwargs("img_size", default=[360,640])[:2])
bgRGB = tuple(get_kwargs("bgRGB", default=(0,0,0,255)))
textRGB = tuple(get_kwargs("textRGB", default=(255,255,255,255)))
mode = get_kwargs("mode", default="RGBA")
margin = get_kwargs("margin", default=10)
fontsize = get_kwargs("fontsize", default=40)
fontwidth = get_kwargs("fontwidth", default=None)
fontheight = get_kwargs("fontheight", default=None)
span = get_kwargs("span", default=1)
audio_path = get_kwargs("audio", default=None)
alpha_range = args.alpha_range
fps = args.fps
spf = 1/fps
kwargs = {
"ttfontname": get_kwargs("ttfontname", default=""),
"img_size": img_size,
"fontsize": fontsize,
"fontwidth": fontwidth,
"fontheight": fontheight,
"margin": margin,
"bgRGB": bgRGB,
"mode": mode,
# "ret_position": "line",
}
text_data = data.get("texts", [[]])
num_texts = len(text_data)
sec_filter = []
for i,page in enumerate(text_data):
start, end = (1e16, 0)
for line in page:
secs = line.get("seconds", [])
start = min(start, *secs)-alpha_range
end = max(end, *secs)
sec_filter.append([i,start,end+alpha_range+span])
duration = max([e[2] for e in sec_filter]) + alpha_range + 1
num_frames = int(duration*fps)
def set_pos(data, name, default):
p = data.get(name, -1)
if p==-1: p = default
return p
def set_fontcolor(start, sec, textRGB, alpha_range):
fc = [0]*4
fc[:3] = textRGB[:3]
fc[3] = min(255, max(0, int((sec-start)/alpha_range*255)))
return tuple(fc)
params = {
"duration" : duration,
"alpha range" : alpha_range,
"FPS": fps,
"span": span,
"audio": audio_path,
}
params.update(kwargs)
print("[Parameters]")
for k,v in params.items():
print(f"* {toACCENT(k)}: {toGREEN(v)}")
root, ext = os.path.splitext(json_path)
video_path = f"{root}_{now_str()}.mp4"
fourcc = cv2.VideoWriter_fourcc('m','p','4','v')
out_video = cv2.VideoWriter(video_path, fourcc, fps, img_size)
monitor = ProgressMonitor(max_iter=num_frames, barname="Editing")
it = sec = 0
while True:
it+=1
bg = Image.new(mode=mode, size=img_size, color=bgRGB)
sec += spf
for i,*_ in filter(lambda x:x[1]<= sec <x[2], sec_filter):
texts = text_data[i]
init_x = x = y = margin
for j,line in enumerate(texts):
words = line.get("words", "")
startSecs = line.get("seconds", [])
x = set_pos(data=line, name="x", default=x)
y = set_pos(data=line, name="y", default=y)
for w,start in zip(words, startSecs):
fc = set_fontcolor(start, sec, textRGB, alpha_range)
bg,(x,_) = draw_text_in_pil(text=w, img=bg, ret_position="word", textRGB=fc, x=x, y=y, **kwargs)
y += fontsize
x = init_x
break
out_video.write(np.asanyarray(bg.convert("RGB")))
monitor.report(it, sec=f"{sec:.2f}/{duration}", text=f"{j+1}/{num_texts}")
if sec>duration:
break
out_video.release()
monitor.remove()
print(f"{toBLUE(video_path)} (No Sound) is created.")
if audio_path is not None:
root, ext = os.path.splitext(audio_path)
if ext!=".mp3":
audio_clip = mp.VideoFileClip(audio_path).subclip()
audio_path = root + ".mp3"
audio_clip.audio.write_audiofile(audio_path)
print(f"Audio file ({toBLUE(audio_path)}) is created.")
video_with_audio_path = f"_audio".join(os.path.splitext(video_path))
clip = mp.VideoFileClip(video_path).subclip()
clip.write_videofile(
video_with_audio_path,
audio=audio_path,
codec='libx264',
audio_codec='aac',
temp_audiofile='temp-audio.m4a',
remove_temp=True
)
command = [
"ffmpeg", "-y",
"-i", video_path,
"-i", audio_path,
"-c:v", "copy",
"-c:a", "copy",
video_with_audio_path,
]
with open("ffmpeg.log", 'w') as f:
process = subprocess.Popen(command, stderr=f)
print(f"{toBLUE(video_with_audio_path)} (With Sound) is created.")
