diff --git a/tools/i18n/locale_diff.py b/tools/i18n/locale_diff.py
index 5fb505d8..327cb1ed 100644
--- a/tools/i18n/locale_diff.py
+++ b/tools/i18n/locale_diff.py
@@ -3,7 +3,7 @@ import os
from collections import OrderedDict
# dir_path = "./i18n/locale" # The path to the i18n locale directory, you can change it to your own path
-dir_path = "./Inference/i18n/locale"
+dir_path = "./tools/srt_slicer/i18n/locale"
# Define the standard file name
standard_file = os.path.join(dir_path, "zh_CN.json")
diff --git a/tools/srt_slicer/i18n/locale/en_US.json b/tools/srt_slicer/i18n/locale/en_US.json
index b1bf423e..415d36d2 100644
--- a/tools/srt_slicer/i18n/locale/en_US.json
+++ b/tools/srt_slicer/i18n/locale/en_US.json
@@ -1,9 +1,13 @@
{
+ "List 合并小工具": "List Merge Tool",
"SRT合并切分插件": "SRT Merge and Split Plugin",
"SRT文件": "SRT File",
+ "SRT编辑界面": "SRT Edit Interface",
"srt文件内容": "SRT File Content",
"上传SRT文件": "Upload SRT File",
"上传文件": "Upload Files",
+ "两个文件夹不能相同！！！": "The two folders cannot be the same!!!",
+ "主文件夹": "Main Folder",
"作者: ": "Author: ",
"使用方法": "How to Use",
"保存合并后字幕": "Save Merged Subtitles",
@@ -12,15 +16,19 @@
"允许最短长度": "Minimum Allowed Length",
"内容预览": "Content Preview",
"切分与保存": "Split and Save",
+ "切分完成": "Split Completed",
"切分并保存音频、list": "Split and Save Audio, List",
"切分预览": "Split Preview",
"判定为短间隔时长": "Judged as Short Interval Duration",
+ "到": " to ",
"前置保留时间": "Preceding Retention Time",
"前置添加静音时间": "Prepend Silence Time",
"句末加句号": "Add Period at the End of Sentence",
"合并后srt文本": "Merged SRT Text",
+ "合并后的List": "Merged List",
"合并字幕": "Merge Subtitles",
"合并字幕设置": "Subtitle Merge Settings",
+ "合并文件夹与List": "Merge Folder and List",
"后置保留时间": "Following Retention Time",
"后置添加静音时间": "Append Silence Time",
"扫描文件夹": "Scan Folder",
@@ -33,8 +41,11 @@
"最大间隔时间": "Maximum Interval Time",
"最长允许单句长度": "Maximum Allowed Sentence Length",
"根据面板合并短句并过滤你不希望出现的句子。": "Merge short sentences according to the panel and filter out sentences you do not want to appear.",
+ "次文件夹": "Second Folder",
+ "正在切分音频": "Splitting Audio",
"正在建设，敬请期待": "Under Construction, Stay Tuned",
"注意：该文件夹已存在": "Warning: The folder already exists",
+ "角色名称，留空使用主文件夹的": "Role Name, Leave Blank to Use Main Folder's",
"语言": "Language",
"读取文件": "Read File",
"读取本地文件": "Read Local File",
diff --git a/tools/srt_slicer/i18n/locale/zh_CN.json b/tools/srt_slicer/i18n/locale/zh_CN.json
index e235eab3..27567f00 100644
--- a/tools/srt_slicer/i18n/locale/zh_CN.json
+++ b/tools/srt_slicer/i18n/locale/zh_CN.json
@@ -1,9 +1,13 @@
{
+ "List 合并小工具": "List 合并小工具",
"SRT合并切分插件": "SRT合并切分插件",
"SRT文件": "SRT文件",
+ "SRT编辑界面": "SRT编辑界面",
"srt文件内容": "srt文件内容",
"上传SRT文件": "上传SRT文件",
"上传文件": "上传文件",
+ "两个文件夹不能相同！！！": "两个文件夹不能相同！！！",
+ "主文件夹": "主文件夹",
"作者: ": "作者: ",
"使用方法": "使用方法",
"保存合并后字幕": "保存合并后字幕",
@@ -12,15 +16,19 @@
"允许最短长度": "允许最短长度",
"内容预览": "内容预览",
"切分与保存": "切分与保存",
+ "切分完成": "切分完成",
"切分并保存音频、list": "切分并保存音频、list",
"切分预览": "切分预览",
"判定为短间隔时长": "判定为短间隔时长",
+ "到": "到",
"前置保留时间": "前置保留时间",
"前置添加静音时间": "前置添加静音时间",
"句末加句号": "句末加句号",
"合并后srt文本": "合并后srt文本",
+ "合并后的List": "合并后的List",
"合并字幕": "合并字幕",
"合并字幕设置": "合并字幕设置",
+ "合并文件夹与List": "合并文件夹与List",
"后置保留时间": "后置保留时间",
"后置添加静音时间": "后置添加静音时间",
"扫描文件夹": "扫描文件夹",
@@ -33,8 +41,11 @@
"最大间隔时间": "最大间隔时间",
"最长允许单句长度": "最长允许单句长度",
"根据面板合并短句并过滤你不希望出现的句子。": "根据面板合并短句并过滤你不希望出现的句子。",
+ "次文件夹": "次文件夹",
+ "正在切分音频": "正在切分音频",
"正在建设，敬请期待": "正在建设，敬请期待",
"注意：该文件夹已存在": "注意：该文件夹已存在",
+ "角色名称，留空使用主文件夹的": "角色名称，留空使用主文件夹的",
"语言": "语言",
"读取文件": "读取文件",
"读取本地文件": "读取本地文件",
diff --git a/tools/srt_slicer/srt_utils.py b/tools/srt_slicer/srt_utils.py
index 6dd226d6..1f030d23 100644
--- a/tools/srt_slicer/srt_utils.py
+++ b/tools/srt_slicer/srt_utils.py
@@ -1,4 +1,5 @@
import srt
+import shutil
def parse_srt_with_lib(content):
@@ -87,10 +88,47 @@ def slice_audio_with_lib(audio_path, save_folder, format, subtitles, pre_preserv
try:
audio = pydub.AudioSegment.from_file(audio_path)
sliced_audio = audio[int(start * 1000):int(end * 1000)]
- file_name = f'{i + 1:03d}.{format}'
+ file_name = f'{character}_{i + 1:03d}.{format}'
save_path = os.path.join(save_folder, file_name)
sliced_audio.export(save_path, format=format)
f.write(f"{file_name}|{character}|{language}|{subtitle.content}\n")
except Exception as e:
raise e
-
\ No newline at end of file
+
+def merge_list_folders(first_list_file, second_list_file, character, first_folder, second_folder):
+ merged_lines = []
+ character1 = ""
+ filenames = set()
+ with open(first_list_file, 'r', encoding="utf-8") as f:
+ first_list = f.readlines()
+ for line in first_list:
+ filename, character1, language, content = line.split('|')
+ filenames.add(filename)
+ if character=="" or character is None:
+ character = character1
+ new_line = f"{filename}|{character}|{language}|{content}"
+ merged_lines.append(new_line)
+ with open(second_list_file, 'r', encoding="utf-8") as f:
+ second_list = f.readlines()
+ for line in second_list:
+ filename, _, language, content = line.split('|')
+ orig_filename = filename
+ num = 1
+ while filename in filenames:
+ filename = f"{filename.rsplit('.', 1)[0]}_{num}.{filename.rsplit('.', 1)[1]}"
+ num += 1
+ try:
+ os.rename(os.path.join(second_folder, orig_filename), os.path.join(first_folder, filename))
+ except Exception as e:
+ raise e
+ new_line = f"{filename}|{character}|{language}|{content}"
+ merged_lines.append(new_line)
+ os.remove(second_list_file)
+ if not os.listdir(second_folder):
+ os.rmdir(second_folder)
+ with open(first_list_file, 'w', encoding="utf-8") as f:
+ f.writelines(merged_lines)
+ return "\n".join(merged_lines)
+
+
+
\ No newline at end of file
diff --git a/tools/srt_slicer/webui.py b/tools/srt_slicer/webui.py
index 436a0484..bd586a6d 100644
--- a/tools/srt_slicer/webui.py
+++ b/tools/srt_slicer/webui.py
@@ -1,9 +1,19 @@
import gradio as gr
import sys
+import os
+sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
+sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '../..')))
+sys.path.append(os.path.abspath(os.path.dirname(__file__)))
sys.path.append('.')
-sys.path.append('..')
-from tools.srt_slicer.srt_utils import merge_subtitles_with_lib, parse_srt_with_lib, generate_srt_with_lib, slice_audio_with_lib, count_words_multilang
+from srt_utils import (
+ merge_subtitles_with_lib,
+ parse_srt_with_lib,
+ generate_srt_with_lib,
+ slice_audio_with_lib,
+ count_words_multilang,
+ merge_list_folders
+)
from i18n.i18n import I18nAuto
@@ -19,7 +29,6 @@ def merge_srt(input_text, output_text, short_interval=0.1, max_interval=1, max_t
return output_text
-
def slice_audio(
input_audio,
save_folder,
@@ -46,6 +55,7 @@ def slice_audio(
os.makedirs(character_folder, exist_ok=True)
subtitles = parse_srt_with_lib(output_text)
try:
+ gr.Info(f"{i18n('正在切分音频')} {input_audio} {i18n('到')} {character_folder}")
slice_audio_with_lib(
input_audio,
character_folder,
@@ -58,9 +68,10 @@ def slice_audio(
language,
character,
)
+ gr.Info(f"{i18n('切分完成')} ")
except Exception as e:
gr.Warning(f"Can't Slice, Error: {e}")
-
+
def get_relative_path(path, base):
return os.path.relpath(path, base)
@@ -124,13 +135,13 @@ def load_srt_from_file(srt_file):
return f.read()
except:
return ""
-
+
def load_audio_from_file(audio_file):
try:
return gr.Audio(audio_file)
except:
return gr.Audio(value=None)
-
+
def load_from_dropdown(input_folder, srt_files_list, audio_files_list):
if isinstance(srt_files_list, str) and isinstance(audio_files_list, str):
srt_file= os.path.join(input_folder, srt_files_list)
@@ -151,7 +162,40 @@ def save_srt_to_file(srt_text, save_folder, character):
srt_file = os.path.join(character_folder, "merged.srt")
with open(srt_file, "w", encoding="utf-8") as f:
f.write(srt_text)
+
+def scan_list_folders(folder):
+ if not os.path.exists(folder):
+ os.makedirs(folder, exist_ok=True)
+ list_folders = []
+ for list_folder in os.listdir(folder):
+ if os.path.isdir(os.path.join(folder, list_folder)):
+ list_folders.append(get_relative_path(os.path.join(folder, list_folder), folder))
+ first_list_folder = ""
+ second_list_folder = ""
+ if len(list_folders) > 0:
+ first_list_folder = second_list_folder = list_folders[0]
+ if len(list_folders) > 1:
+ second_list_folder = list_folders[1]
+ return gr.Dropdown(list_folders, value=first_list_folder), gr.Dropdown(list_folders, value=second_list_folder)
+
+def preview_merged_list(first_list_folder, second_list_folder, merge_list_character_name, save_folder):
+ if first_list_folder == "" or second_list_folder == "":
+ return ""
+ if first_list_folder == second_list_folder:
+ gr.Warning(i18n("两个文件夹不能相同！！！"))
+ return ""
+ first_list_folder = os.path.join(save_folder, first_list_folder)
+ second_list_folder = os.path.join(save_folder, second_list_folder)
+ print(f"first_list_folder: {first_list_folder}, second_list_folder: {second_list_folder}")
+ first_list = os.path.join(first_list_folder, [file for file in os.listdir(first_list_folder) if file.lower().endswith(".list")][0])
+ second_list = os.path.join(second_list_folder, [file for file in os.listdir(second_list_folder) if file.lower().endswith(".list")][0])
+ try:
+ return merge_list_folders(first_list, second_list, merge_list_character_name, first_list_folder, second_list_folder)
+ except Exception as e:
+ gr.Warning(f"Can't Merge, Error: {e}")
+ return ""
+
from datetime import datetime
def change_character_name(input_audio):
@@ -178,63 +222,79 @@ with gr.Blocks() as app:
{i18n("根据面板合并短句并过滤你不希望出现的句子。")}
{i18n("随后保存成切分好的音频与list文件。")}
""")
-
-
- with gr.Row():
- with gr.Column(scale=2) as input_col:
- with gr.Tabs():
- with gr.Tab(i18n("读取本地文件")):
- input_folder = gr.Textbox("input/srt_and_audios", label=i18n("文件夹路径"),interactive=True)
- scan_button = gr.Button(i18n("扫描文件夹"), variant="secondary",interactive=True)
- srt_files_list = gr.Dropdown([], label=i18n("SRT文件"),interactive=True)
- audio_files_list = gr.Dropdown([], label=i18n("音频文件"),interactive=True)
- srt_read_button = gr.Button(i18n("读取文件"), variant="secondary",interactive=True)
- with gr.Tab(i18n("上传文件")):
- input_srt_file = gr.File(label=i18n("上传SRT文件"), type="filepath", file_types=["srt"])
- upload_audio = gr.Audio(type="filepath",label=i18n("音频文件"))
- # input_audio_file = gr.File(label=i18n("上传音频文件"), type="audio", file_types=["mp3", "wav", "ogg"])
- with gr.Tabs():
- with gr.Tab(i18n("内容预览")):
- input_audio = gr.Textbox("", label=i18n("音频文件"),interactive=False)
- input_text = gr.Textbox("", lines=20, max_lines=30, label=i18n("srt文件内容"))
- input_srt_file.change(load_srt_from_file, [input_srt_file], [input_text])
- with gr.Column(scale=1) as control_col:
- with gr.Tabs():
- with gr.Tab(i18n("合并字幕设置")):
- merge_zero_interval = gr.Checkbox(label=i18n("提前合并时间间隔很短的字幕"),interactive=True, value=True)
- short_interval = gr.Slider(value=0.05, minimum=0, maximum=0.5, step=0.005, label=i18n("判定为短间隔时长"),interactive=True,visible=True)
- max_interval = gr.Slider(value=0.8, minimum=0.1, maximum=10, step=0.1, label=i18n("最大间隔时间"),interactive=True)
- max_text_length = gr.Slider(value=50,minimum=5,maximum=200,step=1, label=i18n("最长允许单句长度"),interactive=True)
- add_period = gr.Checkbox(label=i18n("句末加句号"),interactive=True, value=True)
- merge_button = gr.Button(i18n("合并字幕"), variant="primary")
+ with gr.Tabs():
+ with gr.Tab(i18n("SRT编辑界面")):
+ with gr.Row():
+ with gr.Column(scale=2) as input_col:
+ with gr.Tabs():
+ with gr.Tab(i18n("读取本地文件")):
+ input_folder = gr.Textbox("input/srt_and_audios", label=i18n("文件夹路径"),interactive=True)
+ scan_button = gr.Button(i18n("扫描文件夹"), variant="secondary",interactive=True)
+ srt_files_list = gr.Dropdown([], label=i18n("SRT文件"),interactive=True)
+ audio_files_list = gr.Dropdown([], label=i18n("音频文件"),interactive=True)
+ srt_read_button = gr.Button(i18n("读取文件"), variant="secondary",interactive=True)
+ with gr.Tab(i18n("上传文件")):
+ input_srt_file = gr.File(label=i18n("上传SRT文件"), type="filepath", file_types=["srt"])
+ upload_audio = gr.Audio(type="filepath",label=i18n("音频文件"))
+ # input_audio_file = gr.File(label=i18n("上传音频文件"), type="audio", file_types=["mp3", "wav", "ogg"])
+ with gr.Tabs():
+ with gr.Tab(i18n("内容预览")):
+ input_audio = gr.Textbox("", label=i18n("音频文件"),interactive=False)
+ input_text = gr.Textbox("", lines=20, max_lines=30, label=i18n("srt文件内容"))
+ input_srt_file.change(load_srt_from_file, [input_srt_file], [input_text])
+ with gr.Column(scale=1) as control_col:
+ with gr.Tabs():
+ with gr.Tab(i18n("合并字幕设置")):
+ merge_zero_interval = gr.Checkbox(label=i18n("提前合并时间间隔很短的字幕"),interactive=True, value=True)
+ short_interval = gr.Slider(value=0.05, minimum=0, maximum=0.5, step=0.005, label=i18n("判定为短间隔时长"),interactive=True,visible=True)
+ max_interval = gr.Slider(value=0.8, minimum=0.1, maximum=10, step=0.1, label=i18n("最大间隔时间"),interactive=True)
+ max_text_length = gr.Slider(value=50,minimum=5,maximum=200,step=1, label=i18n("最长允许单句长度"),interactive=True)
+ add_period = gr.Checkbox(label=i18n("句末加句号"),interactive=True, value=True)
+ merge_button = gr.Button(i18n("合并字幕"), variant="primary")
- with gr.Tab(i18n("过滤设置")):
- min_length = gr.Slider(value=5, minimum=0, maximum=20, step=1, label=i18n("允许最短长度"),interactive=True)
- filter_english = gr.Checkbox(label=i18n("过滤带有英文的"),interactive=True)
- filter_words = gr.Textbox("", label=i18n("过滤词语，一行一个"),lines=5,max_lines=10,interactive=True)
- filter_button = gr.Button(i18n("过滤字幕"), variant="primary",interactive=False)
- with gr.Tab(i18n("切分与保存")):
- with gr.Group():
- pre_preserve_time = gr.Slider(value=0.1, minimum=0, maximum=1, step=0.01, label=i18n("前置保留时间"),interactive=True)
- post_preserve_time = gr.Slider(value=0.1, minimum=0, maximum=1, step=0.01, label=i18n("后置保留时间"),interactive=True)
- pre_silence_time = gr.Slider(value=0.05, minimum=0, maximum=1, step=0.01, label=i18n("前置添加静音时间"),interactive=True,visible=False)
- post_silence_time = gr.Slider(value=0.1, minimum=0, maximum=1, step=0.01, label=i18n("后置添加静音时间"),interactive=True,visible=False)
- with gr.Group():
- language = gr.Dropdown([i18n(i) for i in ["auto", "zh", "en", "ja", "all_zh", "all_ja"]], value="auto", label=i18n("语言"),interactive=True)
- audio_format = gr.Dropdown(["mp3", "wav", "ogg"], value="wav", label=i18n("音频格式"),interactive=True)
- with gr.Group():
- save_folder = gr.Textbox("output/sliced_audio", label=i18n("保存文件夹"),interactive=True)
- character = gr.Textbox("character", label=i18n("保存子文件夹名称"),interactive=True)
- character_warning = gr.Textbox(i18n("注意：该文件夹已存在"), label=i18n("提示"),interactive=False,visible=False)
- save_srt_button = gr.Button(i18n("保存合并后字幕"),variant="secondary",interactive=True)
- slice_audio_button = gr.Button(i18n("切分并保存音频、list"), variant="primary",interactive=False)
+ with gr.Tab(i18n("过滤设置")):
+ min_length = gr.Slider(value=5, minimum=0, maximum=20, step=1, label=i18n("允许最短长度"),interactive=True)
+ filter_english = gr.Checkbox(label=i18n("过滤带有英文的"),interactive=True)
+ filter_words = gr.Textbox("", label=i18n("过滤词语，一行一个"),lines=5,max_lines=10,interactive=True)
+ filter_button = gr.Button(i18n("过滤字幕"), variant="primary",interactive=False)
+ with gr.Tab(i18n("切分与保存")):
+ with gr.Group():
+ pre_preserve_time = gr.Slider(value=0.1, minimum=0, maximum=1, step=0.01, label=i18n("前置保留时间"),interactive=True)
+ post_preserve_time = gr.Slider(value=0.1, minimum=0, maximum=1, step=0.01, label=i18n("后置保留时间"),interactive=True)
+ pre_silence_time = gr.Slider(value=0.05, minimum=0, maximum=1, step=0.01, label=i18n("前置添加静音时间"),interactive=True,visible=False)
+ post_silence_time = gr.Slider(value=0.1, minimum=0, maximum=1, step=0.01, label=i18n("后置添加静音时间"),interactive=True,visible=False)
+ with gr.Group():
+ language = gr.Dropdown([i18n(i) for i in ["auto", "zh", "en", "ja", "all_zh", "all_ja"]], value="auto", label=i18n("语言"),interactive=True)
+ audio_format = gr.Dropdown(["mp3", "wav", "ogg"], value="wav", label=i18n("音频格式"),interactive=True)
+ with gr.Group():
+ save_folder = gr.Textbox("output/sliced_audio", label=i18n("保存文件夹"),interactive=True)
+ character = gr.Textbox("character", label=i18n("保存子文件夹名称"),interactive=True)
+ character_warning = gr.Textbox(i18n("注意：该文件夹已存在"), label=i18n("提示"),interactive=False,visible=False)
+ save_srt_button = gr.Button(i18n("保存合并后字幕"),variant="secondary",interactive=True)
+ slice_audio_button = gr.Button(i18n("切分并保存音频、list"), variant="primary",interactive=False)
- with gr.Column(scale=2) as output_col:
- with gr.Tabs():
- with gr.Tab(i18n("合并后srt文本")):
- output_text = gr.Textbox("", lines=20, max_lines=30, label="Sliced SRT")
- with gr.Tab(i18n("切分预览")):
- gr.Textbox(i18n("正在建设，敬请期待"), label=i18n("提示"),interactive=False)
+ with gr.Column(scale=2) as output_col:
+ with gr.Tabs():
+ with gr.Tab(i18n("合并后srt文本")):
+ output_text = gr.Textbox("", lines=20, max_lines=30, label="Sliced SRT")
+ with gr.Tab(i18n("切分预览")):
+ gr.Textbox(i18n("正在建设，敬请期待"), label=i18n("提示"),interactive=False)
+ with gr.Tab(i18n("List 合并小工具")):
+ with gr.Row():
+ with gr.Column(scale=2):
+ scan_list_folder = gr.Textbox("output/sliced_audio", label=i18n("文件夹路径"),interactive=True)
+ scan_list_button = gr.Button(i18n("扫描文件夹"), variant="secondary")
+ first_list_folder = gr.Dropdown([], label=i18n("主文件夹"),interactive=True)
+ second_list_folder = gr.Dropdown([], label=i18n("次文件夹"),interactive=True)
+ merge_list_character_name = gr.Textbox("", label=i18n("角色名称，留空使用主文件夹的"),interactive=True)
+ merge_list_button = gr.Button(i18n("合并文件夹与List"), variant="primary")
+ with gr.Column(scale=2):
+ list_preview = gr.Textbox("", lines=20, max_lines=30, label=i18n("合并后的List"))
+
+ scan_list_button.click(scan_list_folders, [scan_list_folder], [first_list_folder, second_list_folder])
+ merge_list_button.click(preview_merged_list, [first_list_folder, second_list_folder, merge_list_character_name, scan_list_folder], [list_preview])
+ save_folder.change(lambda x:gr.Textbox(value=x), [save_folder], [scan_list_folder])
+ scan_list_folder.change(lambda x:gr.Textbox(value=x), [scan_list_folder], [save_folder])
scan_button.click(get_srt_and_audio_files, [input_folder], [srt_files_list, audio_files_list])
merge_zero_interval.change(lambda x: gr.update(visible=x), [merge_zero_interval],[short_interval])
srt_files_list.change(change_srt_file, [input_folder, srt_files_list], [audio_files_list])