-
Notifications
You must be signed in to change notification settings - Fork 4
/
make_subtitle.py
256 lines (200 loc) · 6.76 KB
/
make_subtitle.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
import argparse
import itertools
import json
import math
import os
import re
import book_maker
import openai as openai
import pysrt
import requests
from moviepy.audio.io.AudioFileClip import AudioFileClip
from pydub import AudioSegment
from chatgptapi_translator import ChatGPTAPI
from srt_loader import SRTBookLoader
from utils import LANGUAGES
def video_to_audio(video_file, audio_file, work_dir):
file_to_convert = AudioFileClip(work_dir + "\\" + video_file)
file_to_convert.write_audiofile(work_dir + "\\" + audio_file)
file_to_convert.close()
# 将音频文件切割成固定大小的小文件
def split_audio_file(input_file, output_directory, chunk_size):
# 加载音频文件
audio = AudioSegment.from_file(output_directory + "\\" + input_file)
# 获取音频文件时长
duration_in_sec = math.ceil(audio.duration_seconds)
# 计算需要切割成的小文件数量
num_chunks = math.ceil((os.path.getsize(output_directory + "\\" + input_file) / 1024 / 1024) / chunk_size)
# 计算每个小文件的长度
chunk_length = duration_in_sec / num_chunks
# 切割音频文件
for i in range(num_chunks):
start_time = i * chunk_length * 1000 # 转换为毫秒
end_time = start_time + chunk_length * 1000
chunk = audio[start_time:end_time]
# 保存切割后的小文件
output_file = os.path.join(output_directory, f"chunk_{i + 1}.mp3")
chunk.export(output_file, format="mp3")
return num_chunks
def transcribe_single_audio(api_key, audio_file, new_file_path, src_language):
audio_file = open(audio_file, "rb")
openai.api_key = api_key
transcript = openai.Audio.translate("whisper-1", audio_file, response_format="srt", language=src_language)
f = open(new_file_path, "a+", encoding="utf-8")
f.write(transcript)
f.close()
print(new_file_path + " " + "complete")
def get_all_srt(api_key, basic_file_path, src_language, size):
apikeys = itertools.cycle(api_key.split(","))
for i in range(1, size + 1):
random_key = next(apikeys)
transcribe_single_audio(random_key, basic_file_path + "\\chunk_%s.mp3" % str(i),
basic_file_path + "\\chunk_%s.srt" % str(i), src_language)
def translate_all_subtitle(api_key, basic_file_path, language, size):
for i in range(1, size + 1):
translate_subtitle(api_key, basic_file_path, basic_file_path + "\\chunk_%s.srt" % str(i), language, i)
def translate_subtitle(api_key, basic_file_path, file_path, language, size):
srt = pysrt.open(file_path)
for i in srt:
try:
tran_text = translate_turbo(api_key, i.text, language)
i.text = tran_text
srt.save(basic_file_path + "\\chunk_new_%s.srt" % str(size), encoding="utf-8")
except Exception as e:
print(e)
continue
def translate_turbo(api_key, text, language):
openai.api_key = api_key
completion = openai.ChatCompletion.create(
model="gpt-3.5-turbo",
messages=[
{
"role": "user",
"content": f"Please help me translate: `{text}` to {language},no other words",
}
],
)
t_text = (
completion["choices"][0]
.get("message")
.get("content")
.encode("utf8")
.decode()
)
print(t_text)
return t_text
def translate_davinci(api_key, text, language):
api_url = "https://api.openai.com/v1/completions"
headers = {
"Content-Type": "application/json",
"Authorization": f"Bearer {api_key}"
}
data = {
"prompt": f"Please help me translate: `{text}` to {language},no other words",
"model": "text-davinci-003",
"max_tokens": 2048,
"temperature": 1,
"top_p": 1,
}
r = requests.post(api_url, headers=headers, json=data)
print(r.text)
r = r.json()
t_text = (
r["choices"][0]
.get("text")
.encode("utf8")
.decode()
)
return t_text
def joint_srt(work_dir, size, minute):
f = open(work_dir + "\\result.srt", "w", encoding="utf-8")
result_srt = pysrt.open(work_dir + "\\result.srt")
index = 1
for i in range(1, size + 1):
chunk_srt = pysrt.open(work_dir + "\\chunk_%s.srt" % i)
chunk_srt.shift(seconds=(i - 1) * 60 * minute)
for j in chunk_srt:
j.index = index
result_srt.append(j)
index = index + 1
result_srt.save()
def main(api_key, work_dir, video_file, audio_file, src_language, language):
minute = 15
# num_chunks = 2
video_to_audio(video_file, audio_file, work_dir)
num_chunks = split_audio_file(audio_file, work_dir, minute)
get_all_srt(api_key, work_dir, src_language, num_chunks)
joint_srt(work_dir, num_chunks, minute)
translate_model = ChatGPTAPI
e = SRTBookLoader(
work_dir + "\\result.srt",
translate_model,
api_key,
False,
language=language,
)
e.make_bilingual_book()
# translate_all_subtitle(api_key, work_dir, language, num_chunks)
def check_api_key(api_key):
openai.api_key = api_key
completion = openai.Completion.create(
engine="davinci",
prompt="Hello",
temperature=0,
max_tokens=1,
)
print(completion)
print("API key is valid.")
if __name__ == '__main__':
parser = argparse.ArgumentParser()
audio_file_path = "mp3file.mp3"
parser.add_argument(
"--api_key",
dest="api_key",
type=str,
help="OpenAI API Key",
)
parser.add_argument(
"--work_dir",
dest="work_dir",
type=str,
help="work directory,all the file will be saved here",
)
parser.add_argument(
"--video_file",
dest="video_file",
type=str,
help="video_file_path",
)
parser.add_argument(
"--src_language",
dest="src_language",
type=str,
help="src_language",
)
parser.add_argument(
"--dest_language",
dest="dest_language",
type=str,
help="dest_language",
)
parser.add_argument(
"--proxy",
dest="proxy",
type=str,
help="proxy,like http://127.0.0.1:7890",
)
options = parser.parse_args()
api_key = options.api_key
work_dir = options.work_dir
video_file = options.video_file
src_language = options.src_language
dest_language = options.dest_language
proxy = options.proxy
if proxy:
os.environ["http_proxy"] = proxy
os.environ["https_proxy"] = proxy
src_language = LANGUAGES.get(src_language)
if src_language is None:
raise Exception("wrong src_language,please check it")
main(api_key, work_dir, video_file, audio_file_path, src_language, dest_language)