Skip to content

A
accessibility_movie_2
Commit db7d6ee9 authored by 陈晓勇（工程师）'s avatar 陈晓勇（工程师）
Browse files
Options

dev

parent 4733180f
Expand all Hide whitespace changes
Inline Side-by-side
Showing with 214 additions and 225 deletions
deploy.bat
View file @ db7d6ee9
@echo on
%1 mshta vbscript:CreateObject("Shell.Application").ShellExecute("cmd.exe","/c %~s0 ::","","runas",1)(window.close) && exit
@REM @echo on
@REM %1 mshta vbscript:CreateObject("Shell.Application").ShellExecute("cmd.exe","/c %~s0 ::","","runas",1)(window.close) && exit
cd /d %~dp0
IF EXIST "%PROGRAMFILES(X86)%" (GOTO 64BIT) ELSE (GOTO 32BIT)
......
detect_with_ocr.py
View file @ db7d6ee9
......@@ -27,7 +27,7 @@ import difflib
import re
from typing import Tuple, Union
from utils import reverse_time_to_seconds
from detect_with_asr import create_sheet, write_to_sheet
from main_window import MainWindow, Element
# 字幕的上下边界
......@@ -381,11 +381,26 @@ def process_video(video_path: str, begin: float, end: float, book_path: str, she
lastConf = conf
def add_to_list(mainWindow: MainWindow, element_type: str, li: list):
st_time_sec, ed_time_sec, subtitle, suggest = li
st_time_sec, ed_time_sec = str(st_time_sec), str(ed_time_sec)
# 默认使用配置文件中的语速
speed = mainWindow.projectContext.speaker_speed
aside_head_time = float(reverse_time_to_seconds(mainWindow.aside_head_time)) if mainWindow.aside_head_time != None else float(0)
st_time_sec, ed_time_sec, subtitle, suggest = li
print(">>>>>>>>start time:")
print(st_time_sec)
if (st_time_sec != None and st_time_sec != "" and aside_head_time > float(st_time_sec)):
print(">>>>>>need del")
print(st_time_sec)
if not mainWindow.add_head_aside:
new_element = Element('0.00', "", "", "0/100", "",speed)
mainWindow.projectContext.aside_list.append(new_element)
mainWindow.projectContext.all_elements.append(mainWindow.projectContext.aside_list[-1])
mainWindow.last_aside_index = len(mainWindow.projectContext.all_elements) - 1
mainWindow.add_head_aside = True
return
st_time_sec, ed_time_sec = str(st_time_sec), str(ed_time_sec)
aside = ""
i = len(mainWindow.projectContext.all_elements)
if element_type == "字幕":
......@@ -393,12 +408,19 @@ def add_to_list(mainWindow: MainWindow, element_type: str, li: list):
new_element.print_self()
if mainWindow.last_aside_index != None and mainWindow.projectContext.all_elements[mainWindow.last_aside_index].ed_time_sec == "" and new_element.ed_time_sec != "":
mainWindow.projectContext.all_elements[mainWindow.last_aside_index].ed_time_sec = new_element.st_time_sec
if float(mainWindow.projectContext.all_elements[mainWindow.last_aside_index].ed_time_sec) - float(mainWindow.projectContext.all_elements[mainWindow.last_aside_index].st_time_sec) < 2:
# print(">>>>>>>>>>>remove short aside")
mainWindow.projectContext.aside_list.remove(mainWindow.projectContext.all_elements[mainWindow.last_aside_index])
mainWindow.projectContext.all_elements.remove(mainWindow.projectContext.all_elements[mainWindow.last_aside_index])
mainWindow.last_aside_index = None
mainWindow.projectContext.subtitle_list.append(new_element)
mainWindow.projectContext.all_elements.append(mainWindow.projectContext.subtitle_list[-1])
else:
if i == 0:
st_time_sec = "0.01"
else:
if mainWindow.projectContext.all_elements[i - 1].ed_time_sec == "":
return
st_time_sec = "%.2f" % (float(mainWindow.projectContext.all_elements[i - 1].ed_time_sec) + 0.01)
# 因为暂时没有用到ed_time_sec，所以直接赋值空吧
......@@ -407,6 +429,14 @@ def add_to_list(mainWindow: MainWindow, element_type: str, li: list):
new_element.print_self()
if mainWindow.last_aside_index != None and mainWindow.projectContext.all_elements[mainWindow.last_aside_index].ed_time_sec == "" and new_element.ed_time_sec != "":
mainWindow.projectContext.all_elements[mainWindow.last_aside_index].ed_time_sec = new_element.st_time_sec
if float(mainWindow.projectContext.all_elements[mainWindow.last_aside_index].ed_time_sec) - float(mainWindow.projectContext.all_elements[mainWindow.last_aside_index].st_time_sec) < 2:
# print(">>>>>>>>>>>remove short aside")
mainWindow.projectContext.aside_list.remove(mainWindow.projectContext.all_elements[mainWindow.last_aside_index])
mainWindow.projectContext.all_elements.remove(mainWindow.projectContext.all_elements[mainWindow.last_aside_index])
mainWindow.last_aside_index = None
new_element.suggest = "0/" + new_element.suggest
if (st_time_sec != None and st_time_sec != "" and aside_head_time > float(st_time_sec)):
return
mainWindow.projectContext.aside_list.append(new_element)
mainWindow.projectContext.all_elements.append(mainWindow.projectContext.aside_list[-1])
mainWindow.last_aside_index = len(mainWindow.projectContext.all_elements) - 1
......
excel_utils.py
View file @ db7d6ee9
......@@ -4,7 +4,7 @@ import openpyxl
from management import Element
from utils import reverse_time_to_seconds
from utils import reverse_time_to_seconds, get_seconds
def read_xls(file_path):
print("read_xls")
......@@ -28,9 +28,9 @@ def read_xls(file_path):
cell_value = str(cell_value) if cell_value != None else ""
# print(cell_value)
if col_index == 0:
start_time = reverse_time_to_seconds(cell_value) if cell_value != None else ""
start_time = get_seconds(cell_value) if cell_value != None else ""
if col_index == 1:
end_time = reverse_time_to_seconds(cell_value) if cell_value != None else ""
end_time = get_seconds(cell_value) if cell_value != None else ""
if col_index == 2:
subtitle = cell_value if cell_value != None else ""
if col_index == 3:
......@@ -73,9 +73,9 @@ def read_xlsx(file_path):
cell_value = str(cell_value) if cell_value != None else ""
# print(cell_value)
if col_index == 0:
start_time = reverse_time_to_seconds(cell_value) if cell_value != None else ""
start_time = get_seconds(cell_value) if cell_value != None else ""
if col_index == 1:
end_time = reverse_time_to_seconds(cell_value) if cell_value != None else ""
end_time = get_seconds(cell_value) if cell_value != None else ""
if col_index == 2:
subtitle = cell_value if cell_value != None else ""
if col_index == 3:
......@@ -102,8 +102,10 @@ def checkLength(elements):
for element in elements:
# if int("".join(filter(str.isdigit, element.suggest))) < len(element.aside):
# raise Exception("旁白字数没有按照推荐要求")
if int(element.suggest) < len(element.aside):
if int(element.suggest.split("/")[1]) < len(element.aside):
raise Exception("旁白字数没有按照推荐要求")
else:
element.suggest = str(len(element.aside)) + "/" + element.suggest.split("/")[1]
# elements = read_xlsx("C:/Users/AIA/Desktop/1/121/1.xlsx")
# print(len(elements))
......
main_window.py
View file @ db7d6ee9
This diff is collapsed.
main_window_ui.py
View file @ db7d6ee9
......@@ -334,6 +334,7 @@ class Ui_MainWindow(object):
self.all_tableWidget.setObjectName("all_tableWidget")
self.all_tableWidget.setColumnCount(0)
self.all_tableWidget.setRowCount(0)
# self.all_tableWidget.setStyleSheet(f"QTableWidget::item:selected:enabled {{ background-color: blue; color: white; }}")
self.all_tableWidget.setSelectionBehavior(QtWidgets.QAbstractItemView.SelectRows)
self.horizontalLayout_4.addWidget(self.all_tableWidget)
self.tabWidget.addTab(self.all_tab, "")
......@@ -459,6 +460,8 @@ class Ui_MainWindow(object):
self.menu.setObjectName("menu")
self.menu_2 = QtWidgets.QMenu(self.menubar)
self.menu_2.setObjectName("menu_2")
self.menu_7 = QtWidgets.QMenu(self.menubar)
self.menu_7.setObjectName("menu_7")
# self.menu_3 = QtWidgets.QMenu(self.menubar)
# self.menu_3.setObjectName("menu_3")
self.menu_4 = QtWidgets.QMenu(self.menubar)
......@@ -499,14 +502,16 @@ class Ui_MainWindow(object):
self.action_4.setEnabled(False)
self.action_5 = QtWidgets.QAction("旁白导入",self,triggered=self.import_excel)
self.action_5.setEnabled(False)
self.action_6 = QtWidgets.QAction("字幕上边界++",self,triggered=self.up_ocr)
self.action_6 = QtWidgets.QAction("字幕上边界上移",self,triggered=self.up_ocr)
self.action_6.setEnabled(True)
self.action_7 = QtWidgets.QAction("字幕上边界--",self,triggered=self.down_ocr)
self.action_7 = QtWidgets.QAction("字幕上边界下移",self,triggered=self.down_ocr)
self.action_7.setEnabled(True)
self.action_8 = QtWidgets.QAction("字幕下边界++",self,triggered=self.up_ocr_bottom)
self.action_8 = QtWidgets.QAction("字幕下边界上移",self,triggered=self.up_ocr_bottom)
self.action_8.setEnabled(True)
self.action_9 = QtWidgets.QAction("字幕下边界--",self,triggered=self.down_ocr_bottom)
self.action_9 = QtWidgets.QAction("字幕下边界下移",self,triggered=self.down_ocr_bottom)
self.action_9.setEnabled(True)
self.action_10 = QtWidgets.QAction("片头旁白定位",self,triggered=self.confirm_head_aside)
self.action_10.setEnabled(True)
# self.action_3.setObjectName("action_3")
# self.action_4 = QtWidgets.QAction(MainWindow)
......@@ -542,13 +547,16 @@ class Ui_MainWindow(object):
# self.menu_3.addSeparator()
self.menubar.addAction(self.menu.menuAction())
self.menubar.addAction(self.menu_2.menuAction())
self.menubar.addAction(self.menu_7.menuAction())
self.menubar.addAction(self.action_3)
self.menubar.addAction(self.action_4)
self.menubar.addAction(self.action_5)
self.menubar.addAction(self.action_6)
self.menubar.addAction(self.action_7)
self.menubar.addAction(self.action_8)
self.menubar.addAction(self.action_9)
self.menu_7.addAction(self.action_6)
self.menu_7.addAction(self.action_7)
self.menu_7.addAction(self.action_8)
self.menu_7.addAction(self.action_9)
self.menu_7.addAction(self.action_10)
# self.menubar.addAction(self.menu_5.menuAction())
# self.menubar.addAction(self.menu_6.menuAction())
# self.menubar.addAction(self.menu_3.menuAction())
......@@ -572,6 +580,7 @@ class Ui_MainWindow(object):
self.pb_label.setText(_translate("MainWindow", "刻度"))
self.menu.setTitle(_translate("MainWindow", "文件"))
self.menu_2.setTitle(_translate("MainWindow", "编辑"))
self.menu_7.setTitle(_translate("MainWindow", "旁白检测准备"))
# self.menu_3.setTitle(_translate("MainWindow", "功能按键"))
self.menu_4.setTitle(_translate("MainWindow", "旁白区间检测"))
self.menu_5.setTitle(_translate("MainWindow", "旁白音频合成"))
......
management.py
View file @ db7d6ee9
......@@ -9,7 +9,7 @@ import openpyxl
import constant
from openpyxl.styles import PatternFill, Alignment
from utils import replace_path_suffix, transfer_second_to_time, reverse_time_to_seconds
from utils import replace_path_suffix, transfer_second_to_time, reverse_time_to_seconds, get_seconds, transfer_second_to_all_time
from speech_synthesis import Speaker
class RunThread(threading.Thread):
"""复写线程类，用于解决主线程无法捕捉子线程中异常的问题
......@@ -186,7 +186,7 @@ class ProjectContext:
self.excel_path = info["excel_path"]
self.speaker_info = info["speaker_info"]["speaker_id"]
self.speaker_speed = info["speaker_info"]["speaker_speed"]
self.speaker_type = info["speaker_info"]["speaker_type"] if "speaker_type" in info["speaker_info"] else "科大讯飞"
self.speaker_type = info["speaker_info"]["speaker_type"] if "speaker_type" in info["speaker_info"] else "浙大内部tts"
self.detected = info["detection_info"]["detected"]
self.nd_process = info["detection_info"]["nd_process"]
self.last_time = info["detection_info"]["last_time"]
......@@ -232,7 +232,6 @@ class ProjectContext:
# 先备份文件，再覆盖主文件，可选是否需要备份，默认需要备份
# 20221030：添加旁白检测的进度
def save_project(self, need_save_new: bool=False) -> str:
print("22222sava")
self.save_conf()
# all_element = sorted(all_element, key=lambda x: float(x.st_time_sec))
print("current excel_path:", self.excel_path)
......@@ -261,7 +260,7 @@ class ProjectContext:
def refresh_element(self, row, aside: str):
self.all_elements[int(row)].aside = aside
if not self.initial_ing:
if not self.initial_ing:
save_excel_to_path(self.all_elements, self.excel_path, self.write_header, self.excel_sheet_name)
def refresh_speed(self, row, speed: str)->None:
......@@ -375,8 +374,9 @@ class ProjectContext:
for speaker in content["speaker_zju_details"]:
speaker_name.append(
"，".join([speaker["name"], speaker["gender"], speaker["age_group"]]))
if self.speaker_info is None:
if self.speaker_info is None or self.speaker_info == "":
self.speaker_info = speaker_name[0]
print(">>>>>>>>>>>>>>>>>get all info :" + self.speaker_info)
return tuple(speaker_name)
def init_speakers(self):
......@@ -457,9 +457,23 @@ def write_to_sheet(path: str, sheet_name: str, valuelist: list):
value = ["" if x == None else x for x in value]
# value.insert(1, transfer_second_to_time(value[0])) if value[0] != "" else value.insert(1, "")
# value.insert(3, transfer_second_to_time(value[2])) if value[2] != "" else value.insert(3, "")
value[0] = transfer_second_to_time(value[0]) if value[0] != "" else ""
value[1] = transfer_second_to_time(value[1]) if value[1] != "" else ""
# value[0] = get_seconds(value[0]) if value[0] != "" else ""
# value[1] = get_seconds(value[1]) if value[1] != "" else ""
value[0] = transfer_second_to_all_time(value[0]) if value[0] != "" else ""
value[1] = transfer_second_to_all_time(value[1]) if value[1] != "" else ""
suggest = value[3]
print(">>>>>>>>>>>>>>>>>>>suggest:"+suggest)
if suggest != None and suggest != "":
arrays = suggest.split("/")
if len(arrays) == 2:
value[3] = str(len(value[4])) + "/" + arrays[1]
print(">>>>>>>>>>>>v3:" + value[3])
else:
value[3] = str(len(value[4])) + "/" + arrays[0]
print(">>>>>>>>>>>>v3:" + value[3])
index = len(value)
cur_row = sheet.max_row
for j in range(0, index):
......
myvideoslider.py
View file @ db7d6ee9
......@@ -13,3 +13,4 @@ class myVideoSlider(QSlider):
# self.setValue(int(value)/9)
value = round(value/self.width()*self.maximum()) # 根据鼠标点击的位置和slider的长度算出百分比
self.ClickedValue.emit(value)
self.setFocus()
\ No newline at end of file
res/conf.ini
View file @ db7d6ee9
{"video_path": null, "excel_path": null, "detection_info": {"detected": false, "nd_process": 0.0, "last_time": 0.0, "caption_boundings": [], "has_subtitle": true}, "speaker_info": {"speaker_type": "\u6d59\u5927\u5185\u90e8tts", "speaker_id": "eagle\uff0c\u5973\uff0c\u5e74\u8f7b\u4eba", "speaker_speed": "1.00(4\u5b57/\u79d2)"}}
\ No newline at end of file
{"video_path": null, "excel_path": null, "detection_info": {"detected": false, "nd_process": 0.0, "last_time": 0.0, "caption_boundings": [], "has_subtitle": true}, "speaker_info": {"speaker_type": "", "speaker_id": "eagle\uff0c\u5973\uff0c\u5e74\u8f7b\u4eba", "speaker_speed": "1.00(4\u5b57/\u79d2)"}}
\ No newline at end of file
res/speakers.json
View file @ db7d6ee9
{
"speaker_details": [{
"id": 0,
"name": "晓辰",
"language": "中文（普通话，简体）",
"age_group": "年轻人",
"gender": "女",
"description": "休闲、放松的语音，用于自发性对话和会议听录。",
"audio_path": "./res/speaker_audio/Xiaochen.wav",
"speaker_code": "zh-CN-XiaochenNeural"
},
{
"id": 1,
"name": "晓涵",
"language": "中文（普通话，简体）",
"age_group": "年轻人",
"gender": "女",
"description": "温暖、甜美、富有感情的声音，可用于许多对话场景。",
"audio_path": "./res/speaker_audio/Xiaohan.wav",
"speaker_code": "zh-CN-XiaohanNeural"
},
{
"id": 2,
"name": "晓墨",
"language": "中文（普通话，简体）",
"age_group": "年轻人",
"gender": "女",
"description": "清晰、放松的声音，具有丰富的角色扮演和情感，适合音频书籍。",
"audio_path": "./res/speaker_audio/Xiaomo.wav",
"speaker_code": "zh-CN-XiaomoNeural"
},
{
"id": 7,
"name": "晓晓",
"language": "中文（普通话，简体）",
"age_group": "年轻人",
"gender": "女",
"description": "活泼、温暖的声音，具有多种场景风格和情感。",
"audio_path": "./res/speaker_audio/Xiaoxiao.wav",
"speaker_code": "zh-CN-XiaoxiaoNeural"
},
{
"id": 8,
"name": "晓萱",
"language": "中文（普通话，简体）",
"age_group": "年轻人",
"gender": "女",
"description": "自信、有能力的声音，具有丰富的角色扮演和情感，适合音频书籍。",
"audio_path": "./res/speaker_audio/Xiaoxuan.wav",
"speaker_code": "zh-CN-XiaoxuanNeural"
},
{
"id": 9,
"name": "晓颜",
"language": "中文（普通话，简体）",
"age_group": "年轻人",
"gender": "女",
"description": "训练有素、舒适的语音，用于客户服务和对话场景。",
"audio_path": "./res/speaker_audio/Xiaoyan.wav",
"speaker_code": "zh-CN-XiaoyanNeural"
},
{
"id": 3,
"name": "晓秋",
"language": "中文（普通话，简体）",
"age_group": "中年人",
"gender": "女",
"description": "智能、舒适的语音，适合阅读长内容。",
"audio_path": "./res/speaker_audio/Xiaoqiu.wav",
"speaker_code": "zh-CN-XiaoqiuNeural"
},
{
"id": 4,
"name": "晓秋",
"language": "中文（普通话，简体）",
"age_group": "中年人",
"gender": "女",
"description": "智能、舒适的语音，适合阅读长内容。",
"audio_path": "./res/speaker_audio/Xiaoqiu.wav",
"speaker_code": "zh-CN-XiaoqiuNeural"
},
{
"id": 5,
"name": "晓睿",
"language": "中文（普通话，简体）",
"age_group": "老年",
"gender": "女",
"description": "成熟、睿智的声音，具有丰富的情感，适合音频书籍。",
"audio_path": "./res/speaker_audio/Xiaorui.wav",
"speaker_code": "zh-CN-XiaoruiNeural"
},
{
"id": 6,
"name": "晓双",
"language": "中文（普通话，简体）",
"age_group": "儿童",
"gender": "女",
"description": "可爱、愉悦的语音，可应用于许多儿童相关场景。",
"audio_path": "./res/speaker_audio/Xiaoshuang.wav",
"speaker_code": "zh-CN-XiaoshuangNeural"
},
{
"id": 10,
"name": "晓悠",
"language": "中文（普通话，简体）",
"age_group": "儿童",
"gender": "女",
"description": "天使般的清晰声音，可以应用于许多儿童相关场景。",
"audio_path": "./res/speaker_audio/Xiaoyou.wav",
"speaker_code": "zh-CN-XiaoyouNeural"
},
{
"id": 11,
"name": "云希",
"language": "中文（普通话，简体）",
"age_group": "年轻人",
"gender": "男",
"description": "活泼、阳光的声音，具有丰富的情感，可用于许多对话场景。",
"audio_path": "./res/speaker_audio/Yunxi.wav",
"speaker_code": "zh-CN-YunxiNeural"
},
{
"id": 12,
"name": "云扬",
"language": "中文（普通话，简体）",
"age_group": "年轻人",
"gender": "男",
"description": "专业、流利的声音，具有多种场景风格。",
"audio_path": "./res/speaker_audio/Yunyang.wav",
"speaker_code": "zh-CN-YunyangNeural"
},
{
"id": 13,
"name": "云野",
"language": "中文（普通话，简体）",
"age_group": "中年人",
"gender": "男",
"description": "成熟、放松的声音，具有多种情感，适合音频书籍。",
"audio_path": "./res/speaker_audio/Yunye.wav",
"speaker_code": "zh-CN-YunyeNeural"
}
"speaker_details": [
],
"speaker_zju_details": [{
"id": 0,
......
setting_dialog.py
View file @ db7d6ee9
......@@ -39,6 +39,7 @@ class Setting_Dialog(QDialog, Ui_Dialog):
self.comboBox_2.clear()
# todo 把所有说话人都加上来
self.speaker_li = projectContext.get_all_speaker_info()
# self.speaker_li = []
self.speaker_zju_li = projectContext.get_all_speaker_zju_info() #本地tts
self.speed_list_zju = ["1.00(4字/秒)", "1.10(4.5字/秒)", "1.25(5字/秒)", "1.50(6字/秒)", "1.75(7字/秒)", "2.00(8字/秒)", "2.50(10字/秒)"] #本地tts
......@@ -46,7 +47,8 @@ class Setting_Dialog(QDialog, Ui_Dialog):
# self.comboBox.addItem(i)
self.speed_li_2 = ["1.00(4字/秒)", "1.10(4.5字/秒)", "1.25(5字/秒)", "1.50(6字/秒)", "1.75(7字/秒)", "2.00(8字/秒)", "2.50(10字/秒)"]
# self.comboBox_2.addItems(self.speed_li_2)
self.speaker_types = ["科大讯飞", "浙大内部tts"]
# self.speaker_types = ["科大讯飞", "浙大内部tts"]
self.speaker_types = ["浙大内部tts"]
self.comboBox_0.addItems(self.speaker_types)
print(projectContext.speaker_type)
if projectContext.speaker_type is None or projectContext.speaker_type == "":
......@@ -54,7 +56,7 @@ class Setting_Dialog(QDialog, Ui_Dialog):
else:
self.comboBox_0.setCurrentIndex(self.speaker_types.index(projectContext.speaker_type))
if self.comboBox_0.currentIndex() ==0: #讯飞
if self.comboBox_0.currentIndex() !=0: #讯飞
self.comboBox.addItems(self.speaker_li)
self.comboBox_2.addItems(self.speed_li_2)
else:
......@@ -67,13 +69,13 @@ class Setting_Dialog(QDialog, Ui_Dialog):
self.comboBox.setCurrentIndex(0)
else:
print(projectContext.speaker_info)
self.comboBox.setCurrentIndex(self.speaker_li.index(projectContext.speaker_info) if self.comboBox_0.currentIndex() ==0 else self.speaker_zju_li.index(projectContext.speaker_info))
self.comboBox.setCurrentIndex(self.speaker_li.index(projectContext.speaker_info) if self.comboBox_0.currentIndex() !=0 else self.speaker_zju_li.index(projectContext.speaker_info))
print(projectContext.speaker_speed)
if projectContext.speaker_speed is None or projectContext.speaker_speed == "":
self.comboBox_2.setCurrentIndex(0)
else:
self.comboBox_2.setCurrentIndex(self.speed_li_2.index(projectContext.speaker_speed) if self.comboBox_0.currentIndex() ==0 else self.speed_list_zju.index(projectContext.speaker_speed))
self.comboBox_2.setCurrentIndex(self.speed_li_2.index(projectContext.speaker_speed) if self.comboBox_0.currentIndex() !=0 else self.speed_list_zju.index(projectContext.speaker_speed))
finally:
self.refresh_flag = False
......@@ -84,7 +86,8 @@ class Setting_Dialog(QDialog, Ui_Dialog):
self.comboBox.clear()
self.comboBox_2.clear()
self.projectContext.speaker_type = self.comboBox_0.currentText()
if self.comboBox_0.currentIndex() ==0:
# if self.comboBox_0.currentIndex() ==0:
if self.comboBox_0.currentIndex() !=0:
print("讯飞")
self.comboBox.addItems(self.speaker_li)
self.comboBox_2.addItems(self.speed_li_2)
......@@ -106,12 +109,12 @@ class Setting_Dialog(QDialog, Ui_Dialog):
if self.projectContext.speaker_info is None or self.projectContext.speaker_info == "" :
self.comboBox.setCurrentIndex(0)
else:
self.comboBox.setCurrentIndex(self.speaker_li.index(self.projectContext.speaker_info) if self.comboBox_0.currentIndex() ==0 else self.speaker_zju_li.index(self.projectContext.speaker_info))
self.comboBox.setCurrentIndex(self.speaker_li.index(self.projectContext.speaker_info) if self.comboBox_0.currentIndex() !=0 else self.speaker_zju_li.index(self.projectContext.speaker_info))
if self.projectContext.speaker_speed is None or self.projectContext.speaker_speed == "":
self.comboBox_2.setCurrentIndex(0)
else:
self.comboBox_2.setCurrentIndex(self.speed_li_2.index(self.projectContext.speaker_speed) if self.comboBox_0.currentIndex() ==0 else self.speed_list_zju.index(self.projectContext.speaker_speed))
self.comboBox_2.setCurrentIndex(self.speed_li_2.index(self.projectContext.speaker_speed) if self.comboBox_0.currentIndex() !=0 else self.speed_list_zju.index(self.projectContext.speaker_speed))
def speaker_change_slot(self):
"""切换说话人
......
speech_synthesis.py
View file @ db7d6ee9
......@@ -23,8 +23,8 @@ from typing import Tuple
import datetime
import numpy as np
from azure.cognitiveservices.speech import SpeechConfig, SpeechSynthesizer, ResultReason, AudioDataStream
from azure.cognitiveservices.speech.audio import AudioOutputConfig
# from azure.cognitiveservices.speech import SpeechConfig, SpeechSynthesizer, ResultReason, AudioDataStream
# from azure.cognitiveservices.speech.audio import AudioOutputConfig
import openpyxl
import shutil
from vits_chinese import tts
......@@ -82,8 +82,11 @@ def choose_speaker(speaker_name: str) -> Speaker:
Returns:
Speaker: 返回对应说话人，如果没有这个说话人则报错
"""
print(">>>>>>>>>>>>>speakerName:" + speaker_name)
for speaker in speakers:
print(">>>>>>>>>>>>>speaker:" + speaker.name)
if speaker.name == speaker_name:
return speaker
raise ValueError
......@@ -106,41 +109,42 @@ def speech_synthesis(text: str, output_file: str, speaker: Speaker, speed: float
if speaker.speaker_type != None and speaker.speaker_type == "1":
tts(text, speed, output_file)
else:
speech_config = SpeechConfig(
subscription="db34d38d2d3447d482e0f977c66bd624",
region="eastus"
)
print("1")
# speech_config = SpeechConfig(
# subscription="db34d38d2d3447d482e0f977c66bd624",
# region="eastus"
# )
speech_config.speech_synthesis_language = "zh-CN"
speech_config.speech_synthesis_voice_name = speaker.speaker_code
# speech_config.speech_synthesis_language = "zh-CN"
# speech_config.speech_synthesis_voice_name = speaker.speaker_code
# 先把合成的语音文件输出得到tmp.wav中，便于可能的调速需求
# # 先把合成的语音文件输出得到tmp.wav中，便于可能的调速需求
synthesizer = SpeechSynthesizer(speech_config=speech_config, audio_config=None)
ssml_string = f"""
<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="{speech_config.speech_synthesis_language}">
<voice name="{speaker.speaker_code}">
<prosody rate="{round((speed - 1.0) * 100, 2)}%">
{text}
</prosody>
</voice>
</speak>"""
result = synthesizer.speak_ssml_async(ssml_string).get()
stream = AudioDataStream(result)
stream.save_to_wav_file(output_file)
print(result.reason)
while result.reason == ResultReason.Canceled:
cancellation_details = result.cancellation_details
print("取消的原因", cancellation_details.reason, cancellation_details.error_details)
time.sleep(1)
synthesizer.stop_speaking()
del synthesizer
synthesizer = SpeechSynthesizer(speech_config=speech_config, audio_config=None)
result = synthesizer.speak_ssml_async(ssml_string).get()
stream = AudioDataStream(result)
stream.save_to_wav_file(output_file)
print(result.reason)
# synthesizer = SpeechSynthesizer(speech_config=speech_config, audio_config=None)
# ssml_string = f"""
# <speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="{speech_config.speech_synthesis_language}">
# <voice name="{speaker.speaker_code}">
# <prosody rate="{round((speed - 1.0) * 100, 2)}%">
# {text}
# </prosody>
# </voice>
# </speak>"""
# result = synthesizer.speak_ssml_async(ssml_string).get()
# stream = AudioDataStream(result)
# stream.save_to_wav_file(output_file)
# print(result.reason)
# while result.reason == ResultReason.Canceled:
# cancellation_details = result.cancellation_details
# print("取消的原因", cancellation_details.reason, cancellation_details.error_details)
# time.sleep(1)
# synthesizer.stop_speaking()
# del synthesizer
# synthesizer = SpeechSynthesizer(speech_config=speech_config, audio_config=None)
# result = synthesizer.speak_ssml_async(ssml_string).get()
# stream = AudioDataStream(result)
# stream.save_to_wav_file(output_file)
# print(result.reason)
# detached
def change_speed_and_volume(wav_path: str, speed: float = 1.0):
......
start.py
View file @ db7d6ee9
......@@ -56,7 +56,20 @@ def change_project_path(path):
if __name__ == '__main__':
try:
# subprocess.call(['deploy.bat'])
if not is_lav_filters_installed() or not is_file_copyed():
# if not is_lav_filters_installed() or not is_file_copyed():
# QCoreApplication.setAttribute(Qt.AA_EnableHighDpiScaling)
# QCoreApplication.setAttribute(Qt.AA_UseHighDpiPixmaps)
# app = QApplication(sys.argv)
# app.setWindowIcon(QIcon("./res/images/eagle_2.ico"))
# apply_stylesheet(app, theme='dark_amber.xml')
# mainWindow = MainWindow(project_path)
# QtWidgets.QMessageBox.critical(mainWindow,'警告','视频解码器未正常安装',QtWidgets.QMessageBox.Yes)
# else:
# QCoreApplication.setAttribute(Qt.AA_EnableHighDpiScaling)
# QCoreApplication.setAttribute(Qt.AA_UseHighDpiPixmaps)
# currentExitCode = MainWindow.EXIT_CODE_REBOOT
if not os.path.exists("C:\LavFilters") and not os.path.exists("C:\Program Files (x86)\LAV Filters"):
QCoreApplication.setAttribute(Qt.AA_EnableHighDpiScaling)
QCoreApplication.setAttribute(Qt.AA_UseHighDpiPixmaps)
app = QApplication(sys.argv)
......
start.spec
View file @ db7d6ee9
......@@ -3,7 +3,7 @@
block_cipher = None
env_dir = 'C:/Users/AIA/.conda/envs/testmovie/Lib/site-packages/'
missingPkgs = ['Microsoft.CognitiveServices.Speech.core.dll', 'decorator.py', 'google', 'paddle', 'paddleocr', 'PIL', 'requests', 'urllib3', 'http', 'idna', 'certifi', 'setuptools', 'astor', 'charset_normalizer']
missingPkgs = ['Microsoft.CognitiveServices.Speech.core.dll', 'decorator.py', 'google', 'paddleocr', 'PIL', 'requests', 'urllib3', 'http', 'idna', 'certifi', 'setuptools', 'astor', 'charset_normalizer']
def add_missing_packages(lst):
pkgs = []
......
utils.py
View file @ db7d6ee9
......@@ -38,6 +38,29 @@ def validate_and_get_filepath(file_info) -> Tuple[str, bool]:
return "", False
return file_info[0][0], True
def get_seconds(time_str: str):
try:
# print(">>>>>>>>>>reverse time")
# print(time_str)
if time_str is None or time_str == "":
return time_str
parts = time_str.split(":")
if len(parts) != 3:
return time_str
hour = int(parts[0])
minutes = int(parts[1])
seconds_parts = parts[2].split(".")
if len(seconds_parts) != 2:
return time_str
seconds = int(seconds_parts[0])
milliseconds = int(seconds_parts[1])
total_seconds = hour * 3600 + minutes * 60 + seconds + milliseconds / 1000
return str(total_seconds)
except Exception as e:
print(e)
return time_str
def trans_to_seconds(timePoint: str) -> float:
"""将用户输入的时间字符串转换为秒数
......@@ -47,17 +70,43 @@ def trans_to_seconds(timePoint: str) -> float:
Returns:
float: 时间字符串对应秒数
"""
time_in_seconds = 0
timePoints = timePoint.split(':')
units = 1
for i in range(len(timePoints) - 1, -1, -1):
time_in_seconds += units * float(timePoints[i])
units *= 60
return time_in_seconds
# time_in_seconds = 0
# timePoints = timePoint.split(':')
# units = 1
# for i in range(len(timePoints) - 1, -1, -1):
# time_in_seconds += units * float(timePoints[i])
# units *= 60
# return time_in_seconds
try:
return float(reverse_time_to_seconds(timePoint))
except Exception as e:
return None
def transfer_second_to_time(sec: str) -> str:
"""将秒数转换为"hh:mm:ss.xxx"格式的时间字符串
Args:
sec (str): 待转换的描述
Returns:
str: "hh:mm:ss.xxx"格式的时间字符串
"""
try:
duration = int(float(sec))
hour = int(duration/3600)
minutes = int((duration % 3600)/60)
seconds = int(duration%60)
# msec = round((float(sec) - hour * 3600 - minutes * 60 - seconds) * 1000)
# time = "%02d:%02d:%02d.%03d" % (hour, minutes, seconds, msec)
time = "%02d:%02d:%02d" % (hour, minutes, seconds)
return time
except Exception as e:
print(e)
return sec
def transfer_second_to_all_time(sec: str) -> str:
"""将秒数转换为"hh:mm:ss.xxx"格式的时间字符串
Args:
sec (str): 待转换的描述
......@@ -71,6 +120,7 @@ def transfer_second_to_time(sec: str) -> str:
seconds = int(duration%60)
msec = round((float(sec) - hour * 3600 - minutes * 60 - seconds) * 1000)
time = "%02d:%02d:%02d.%03d" % (hour, minutes, seconds, msec)
# time = "%02d:%02d:%02d" % (hour, minutes, seconds)
return time
except Exception as e:
print(e)
......@@ -97,13 +147,13 @@ def reverse_time_to_seconds(time_str: str) -> str:
hour = int(parts[0])
minutes = int(parts[1])
seconds_parts = parts[2].split(".")
if len(seconds_parts) != 2:
return time_str
# if len(seconds_parts) != 2:
# return time_str
seconds = int(seconds_parts[0])
milliseconds = int(seconds_parts[1])
# milliseconds = int(seconds_parts[1])
total_seconds = hour * 3600 + minutes * 60 + seconds + milliseconds / 1000
# total_seconds = hour * 3600 + minutes * 60 + seconds + milliseconds / 1000
total_seconds = hour * 3600 + minutes * 60 + seconds
print(str(total_seconds))
return str(total_seconds)
except Exception as e:
......@@ -189,7 +239,8 @@ def get_progress_with_cmd(cmd: str, state=None):
if result is not None:
elapsed_time = result.groupdict()['time']
# 此处可能会出现进度超过100%，未对数值进行纠正
progress = trans_to_seconds(elapsed_time) / trans_to_seconds(duration)
# progress = trans_to_seconds(elapsed_time) / trans_to_seconds(duration)
progress = float(get_seconds(elapsed_time)) / float(get_seconds(duration))
state[0] = pre + progress * 0.2
print(elapsed_time)
print(progress)
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment