Merge remote-tracking branch 'refs/remotes/origin/feat_1' into feat_1

.gitignore

@@ -9,4 +9,8 @@ dist
 build
 log
 product_test
+res/ffmpeg-4.3.1/bin/output.mp4
+res/ffmpeg-4.3.1/bin/qiji_local.mp4
+venv/
+venv37/
 shenming_test
\ No newline at end of file

PaddlePaddle_DeepSpeech2/infer_path.py

@@ -78,7 +78,8 @@ def predict_long_audio_with_paddle(wav_path, pre_time, book_name, sheet_name, st
                     (time_stamps[i][0] + pre_time) * normal_speed)
                 print("插入旁白，推荐字数为%d" % recommend_lens)
                 # narratages.append(["", "", "", "插入旁白，推荐字数为%d" % recommend_lens])
-                write_to_sheet(book_name, sheet_name, ["", "", "", "插入旁白，推荐字数为%d" % recommend_lens])
+                # write_to_sheet(book_name, sheet_name, ["", "", "", "插入旁白，推荐字数为%d" % recommend_lens])
+                write_to_sheet(book_name, sheet_name, ["", "", "", "%d" % recommend_lens])
             # narratages.append([round(time_stamps[i][0] + pre_time, 2), round(time_stamps[i][1] + pre_time, 2),
             # text, ''])
             write_to_sheet(book_name, sheet_name,

create_dialog.py

@@ -16,14 +16,23 @@ from prompt_dialog import Prompt_Dialog
 
 class Create_Dialog(QDialog, Ui_Dialog):
     init_project_signal = pyqtSignal(str)
-    def __init__(self):
+    def __init__(self, mainWindow):
         super(Create_Dialog, self).__init__()
+        self.mainWindow = mainWindow
         self.setupUi(self)
         self.setWindowTitle("新建工程")
         self.get_dir.clicked.connect(self.choose_root)
         self.confirm.clicked.connect(self.create_project)
+        self.get_video_dir.clicked.connect(self.choose_video)
         self.prompt_dialog = Prompt_Dialog()
         self.cancel.clicked.connect(self.close_dialog)
+        self.video_path = None
+
+    def choose_video(self):
+        path = QFileDialog.getOpenFileUrl(self, "选择待导入视频", QUrl(
+            os.getcwd()), "Video Files(*.mp4 *.rmvb *mkv *avi *.);; 所有文件(*.*)")[0]
+        self.video_input.setText(path.path()[1:])
+        self.video_path = path
         
     def choose_root(self):
         """选择存放新工程的文件夹
@@ -54,6 +63,7 @@ class Create_Dialog(QDialog, Ui_Dialog):
             self.prompt_dialog.show_with_msg("请输入工程名称")
         else:
             self.prompt_dialog.show_with_msg("请输入合法文件夹路径")
+        self.mainWindow.play_video(self.video_path)
     
     def close_dialog(self):
         """关闭窗口

create_dialog_ui.py

@@ -29,6 +29,13 @@ class Ui_Dialog(object):
         self.get_dir.setObjectName("get_dir")
         self.gridLayout.addWidget(self.get_dir, 0, 2, 1, 1)
 
+        self.video_input = QtWidgets.QLineEdit(Dialog)
+        self.video_input.setObjectName("video_input")
+        self.gridLayout.addWidget(self.video_input, 2, 1, 1, 1)
+        self.get_video_dir = QtWidgets.QPushButton(Dialog)
+        self.get_video_dir.setObjectName("get_video_dir")
+        self.gridLayout.addWidget(self.get_video_dir, 2, 2, 1, 1)
+
         self.rootLabel = QtWidgets.QLabel(Dialog)
         self.rootLabel.setObjectName("rootLabel")
         self.gridLayout.addWidget(self.rootLabel, 0, 0, 1, 1)
@@ -37,6 +44,10 @@ class Ui_Dialog(object):
         self.nameLabel.setObjectName("nameLabel")
         self.gridLayout.addWidget(self.nameLabel, 1, 0, 1, 1)
 
+        self.videoPath = QtWidgets.QLabel(Dialog)
+        self.videoPath.setObjectName("videoPath")
+        self.gridLayout.addWidget(self.videoPath, 2, 0, 1, 1)
+
         self.gridLayout.addWidget(self.name_input, 1, 1, 1, 1)
         self.gridLayout_2.addLayout(self.gridLayout, 0, 0, 1, 1)
         self.horizontalLayout = QtWidgets.QHBoxLayout()
@@ -68,6 +79,8 @@ class Ui_Dialog(object):
         Dialog.setWindowTitle(_translate("Dialog", "Dialog"))
         self.nameLabel.setText(_translate("Dialog", "工程名称"))
         self.rootLabel.setText(_translate("Dialog", "目标路径"))
+        self.videoPath.setText(_translate("Dialog", "视频路径"))
         self.get_dir.setText(_translate("Dialog", "打开文件夹"))
+        self.get_video_dir.setText(_translate("Dialog", "导入视频"))
         self.confirm.setText(_translate("Dialog", "确认"))
         self.cancel.setText(_translate("Dialog", "取消"))

deploy.bat

-@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)
@@ -16,6 +16,7 @@ GOTO END
 
 :END
 
+c:
 cd C:\LavFilters\
 call install_audio.bat
 call install_splitter.bat

detect_with_asr.py

@@ -98,7 +98,7 @@ def detect_with_asr(video_path: str, book_path: str, start_time=0, end_time=-1,
         os.remove(book_path)
     book_name_xlsx = book_path
     sheet_name_xlsx = "旁白插入位置建议"
-    table_head = [["起始时间", "终止时间", "字幕", '建议', '解说脚本']]
+    table_head = [["起始时间", "终止时间", "字幕", '建议旁白字数', '解说脚本']]
     create_sheet(book_name_xlsx, sheet_name_xlsx, table_head)
 
     sys.path.append("PaddlePaddle_DeepSpeech2")

ding_notify.conf

+{
+    "user_name":"盲文出版社",
+    "dd_url_prefix":"https://oapi.dingtalk.com/robot/send?access_token=a7e6b223bc0a6e4b5e3f64b1570d0c884db0d833592f5e84b020c2896130d659",
+    "secret":"SEC252452789bc17cfb1e6c973f837426fa0f9fd50c8793239c070e9d3c021d8109"
+}
\ No newline at end of file

ding_utils.py

@@ -9,35 +9,44 @@ import base64
 import hashlib
 import urllib.parse
 
-user_name = "盲文出版社"
-dd_url_prefix = "https://oapi.dingtalk.com/robot/send?access_token=a7e6b223bc0a6e4b5e3f64b1570d0c884db0d833592f5e84b020c2896130d659"
-# dd_url_prefix = "https://oapi.dingtalk.com/robot/send?access_token=12a5e64b24e7345fed2415c5c2f9101180763e327b21727cd35fdaa3fb274e74"
-secret = 'SEC252452789bc17cfb1e6c973f837426fa0f9fd50c8793239c070e9d3c021d8109'
-# secret = "SECb5d68b27bd2e4694667d36c0f91336f3641b3287b945906a83997353f961d415"
+with open("./ding_notify.conf", "r", encoding="utf-8") as f:
+    json_obj = json.load(f)
+# user_name = "盲文出版社"
+# dd_url_prefix = "https://oapi.dingtalk.com/robot/send?access_token=a7e6b223bc0a6e4b5e3f64b1570d0c884db0d833592f5e84b020c2896130d659"
+# # dd_url_prefix = "https://oapi.dingtalk.com/robot/send?access_token=12a5e64b24e7345fed2415c5c2f9101180763e327b21727cd35fdaa3fb274e74"
+# secret = 'SEC252452789bc17cfb1e6c973f837426fa0f9fd50c8793239c070e9d3c021d8109'
+# # secret = "SECb5d68b27bd2e4694667d36c0f91336f3641b3287b945906a83997353f961d415"
+user_name = json_obj["user_name"]
+dd_url_prefix = json_obj["dd_url_prefix"]
+secret = json_obj["secret"]
 
 def notify(err):
-    timestamp = str(round(time.time() * 1000))
-    secret_enc = secret.encode('utf-8')
-    string_to_sign = '{}\n{}'.format(timestamp, secret)
-    string_to_sign_enc = string_to_sign.encode('utf-8')
-    hmac_code = hmac.new(secret_enc, string_to_sign_enc, digestmod=hashlib.sha256).digest()
-    sign = urllib.parse.quote_plus(base64.b64encode(hmac_code))
-    print(timestamp)
-    print(sign)
-    dd_url = dd_url_prefix + "&timestamp=" + timestamp + "&sign=" + sign
-    h = {"Content-Type": "application/json"}
-    d = {
-        "msgtype": "markdown",
-        "markdown": {
-            "title": "msg",
-            "text": "| 用户名称:%s \n" % (user_name) +
-                    "\n" +
-                    "| 异常时间:%s \n" % (time.strftime('%Y:%m:%d %H:%M:%S', time.localtime(int(time.time())))) +
-                    "\n" +
-                    "| 异常信息:%s \n" % (err)
-                    
+    try:
+        timestamp = str(round(time.time() * 1000))
+        secret_enc = secret.encode('utf-8')
+        string_to_sign = '{}\n{}'.format(timestamp, secret)
+        string_to_sign_enc = string_to_sign.encode('utf-8')
+        hmac_code = hmac.new(secret_enc, string_to_sign_enc, digestmod=hashlib.sha256).digest()
+        sign = urllib.parse.quote_plus(base64.b64encode(hmac_code))
+        print(timestamp)
+        print(sign)
+        dd_url = dd_url_prefix + "&timestamp=" + timestamp + "&sign=" + sign
+        h = {"Content-Type": "application/json"}
+        d = {
+            "msgtype": "markdown",
+            "markdown": {
+                "title": "msg",
+                "text": "| 用户名称:%s \n" % (user_name) +
+                        "\n" +
+                        "| 异常时间:%s \n" % (time.strftime('%Y:%m:%d %H:%M:%S', time.localtime(int(time.time())))) +
+                        "\n" +
+                        "| 异常信息:%s \n" % (err)
+                        
+            }
         }
-    }
-    r = requests.post(dd_url, json.dumps(d), headers=h)
-    print(r.status_code)
-    print(r.text)
\ No newline at end of file
+        r = requests.post(dd_url, json.dumps(d), headers=h)
+        print(r.status_code)
+        print(r.text)
+    except Exception as e:
+        print("ding notify err")
+        print(e)
\ No newline at end of file

excel_utils.py

@@ -4,6 +4,8 @@ import openpyxl
 
 from management import Element
 
+from utils import reverse_time_to_seconds, get_seconds
+
 def read_xls(file_path):
     print("read_xls")
     elements = []
@@ -26,20 +28,20 @@ def read_xls(file_path):
             cell_value = str(cell_value) if cell_value != None else ""
             # print(cell_value)
             if col_index == 0:
-                start_time = cell_value if cell_value != None else ""
+                start_time = get_seconds(cell_value) if cell_value != None else ""
+            if col_index == 1:
+                end_time = get_seconds(cell_value) if cell_value != None else ""
             if col_index == 2:
-                end_time = cell_value if cell_value != None else ""
-            if col_index == 4:
                 subtitle = cell_value if cell_value != None else ""
-            if col_index == 5:
+            if col_index == 3:
                 if cell_value == None or cell_value == "":
                     break
                 suggest = cell_value if cell_value != None else ""
-            if col_index == 6:
+            if col_index == 4:
                 if cell_value == None or cell_value == "":
                     break
                 aside = cell_value if cell_value != None else ""
-            if col_index == 7:
+            if col_index == 5:
                 speed = cell_value if cell_value != None else ""
             col_index = col_index + 1
         if suggest != "" and aside != "":
@@ -71,33 +73,39 @@ def read_xlsx(file_path):
             cell_value = str(cell_value) if cell_value != None else ""
             # print(cell_value)
             if col_index == 0:
-                start_time = cell_value if cell_value != None else ""
+                start_time = get_seconds(cell_value) if cell_value != None else ""
+            if col_index == 1:
+                end_time = get_seconds(cell_value) if cell_value != None else ""
             if col_index == 2:
-                end_time = cell_value if cell_value != None else ""
-            if col_index == 4:
                 subtitle = cell_value if cell_value != None else ""
-            if col_index == 5:
+            if col_index == 3:
                 if cell_value == None or cell_value == "":
                     break
                 suggest = cell_value if cell_value != None else ""
-            if col_index == 6:
+            if col_index == 4:
                 if cell_value == None or cell_value == "":
                     break
                 aside = cell_value if cell_value != None else ""
-            if col_index == 7:
+            if col_index == 5:
                 speed = cell_value if cell_value != None else ""
             col_index = col_index + 1
-        if suggest != "" and aside != "":
+        if suggest != None and suggest != "" and aside != None and aside != "":
             new_element = Element(start_time, end_time, subtitle, suggest, aside, speed)
-            # new_element.print_self()
+            new_element.print_self()
             elements.append(new_element)
     checkLength(elements)
+    print("element size:")
+    print(len(elements))
     return elements
 
 def checkLength(elements):
     for element in elements:
-        if int("".join(filter(str.isdigit, element.suggest))) < len(element.aside):
+        # 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.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))

ffmpeg_util.py

+import subprocess
+import os
+
+import subprocess
+import math
+import time
+import shutil
+
+def calculate_audio_volume(file_path):
+    # 使用FFmpeg获取音频分贝值
+    command = [
+        'res/ffmpeg-4.3.1/bin/ffmpeg', '-i', file_path,
+        '-af', 'volumedetect', '-f', 'null', '-'
+    ]
+    print(command)
+    result = subprocess.run(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
+    output = result.stderr
+    # 从输出中解析分贝值
+    for line in output.split('\n'):
+        if 'mean_volume' in line:
+            print(line)
+            
+            a = float(line.split(':')[1].strip().split("d")[0].strip())
+            print(a)
+            # mean_volume = float((line.split(':')[1].strip()).split(" ")[0])
+            return a
+
+    return None
+
+# def adjust_audio_volume(input_file, output_file, multiplier):
+#     # 使用FFmpeg将音频音量设置为给定倍数
+#     command = [
+#         'ffmpeg', '-i', input_file,
+#         '-af', f'volume={multiplier}*', output_file
+#     ]
+#     subprocess.run(command)
+
+def adjust_audio_volume(input_file, volume_adjustment):
+    temp_path = "./temp.wav"
+    command = [
+        "res/ffmpeg-4.3.1/bin/ffmpeg",
+        "-i", input_file,
+        "-af", f"volume={volume_adjustment}",
+        "-y",  # 覆盖输出文件，加上此选项
+        temp_path  # 输出文件与输入文件相同
+    ]
+    subprocess.run(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+    os.remove(input_file)
+    # os.rename(temp_path, input_file)
+    shutil.move(temp_path, input_file)
+
+
+def cg_wav(first_file_path, second_file_path, rate):
+    try:
+        print(first_file_path, second_file_path, str(rate))
+        # 计算第一个文件的音量
+        first_file_volume = calculate_audio_volume(first_file_path)
+
+        if first_file_volume is not None:
+            # 计算第二个文件的音量并计算倍数
+            second_file_volume = calculate_audio_volume(second_file_path)
+
+            if second_file_volume is not None:
+                # volume_multiplier = 10 ** ((first_file_volume - second_file_volume) / 20)  # 转换为幅度倍数
+
+                # # 调整第二个文件的音量
+                # adjusted_output_file = 'path/to/output/adjusted_file.wav'
+                # 计算第二个文件应改为自身的大小的倍数
+
+                volume_difference_db = first_file_volume - second_file_volume
+                volume_difference_linear = 10 ** (volume_difference_db / 20)  # 转换为线性倍数
+                adjusted_multiplier = volume_difference_linear * rate
+                adjust_audio_volume(second_file_path, adjusted_multiplier)
+            else:
+                print("无法计算第二个文件的音量")
+        else:
+            print("无法计算第一个文件的音量")
+    except Exception as e:
+        print("cg_wav err")
+        print(e)
+
+
+# cg_wav("C:/Users/Administrator/Desktop/无障碍电影/fuiii0000/output/0.01.wav","C:/Users/Administrator/Desktop/无障碍电影/fuiii0000\output\0.01.wav",0.26)
+# calculate_audio_volume("./0.01.wav")
+# adjust_audio_volume("./0.01.wav",1.335)
+# calculate_audio_volume("./0.01.wav")
+
+# def test(projectContext):
+#     from split_wav import extract_audio
+#     video_path = projectContext.video_path
+#     output_dir = os.path.join(projectContext.project_base_dir, "output")
+#     extract_audio(video_path, output_dir, 0, -1)
+
+# self.projectContext.video_path
+# output_dir = os.path.join(self.projectContext.project_base_dir, "output")
+# files = os.listdir(output_dir):
+#     for i, f in enumerate(files):
+#         fname = '.'.join(f.split('.')[:-1])
+#         try:
+#             st_time = float(fname)
+#             cur_audio, _ = soundfile.read(os.path.join(output_dir, f))
+#             # print(len(cur_audio))
+#             st_index = int(st_time * freq)
+#             audio_len = len(cur_audio)
+#             blank_audio[st_index: st_index + audio_len] = cur_audio
+#             origin_wav[st_index: st_index + audio_len] *= origin_audio_rate
+#             state[0] = float((i + 1) / len(files)) * 0.6 + 0.2
+#         except:
+#             continue
\ No newline at end of file

management.py

@@ -9,7 +9,7 @@ import openpyxl
 import constant
 from openpyxl.styles import PatternFill, Alignment
 
-from utils import replace_path_suffix, transfer_second_to_time
+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
@@ -125,8 +125,9 @@ class ProjectContext:
         self.speaker_speed = None
         self.duration = 0
         # 一些常量
-        self.header = ["起始时间", "终止时间", "字幕", '建议', '解说脚本', "语速"]
-        self.write_header = ["起始时间", "起始时间（转换后）", "终止时间", "终止时间（转换后）", "字幕", '建议', '解说脚本', "语速"]
+        self.header = ["起始时间", "终止时间", "字幕", '建议旁白字数', '解说脚本', "语速"]
+        # self.write_header = ["起始时间", "起始时间（转换后）", "终止时间", "终止时间（转换后）", "字幕", '建议', '解说脚本', "语速"]
+        self.write_header = ["起始时间", "终止时间", "字幕", '建议旁白字数', '解说脚本', "语速"]
         self.aside_header = ['起始时间', '推荐字数', '解说脚本',"语速", "预览音频"]
         self.subtitle_header = ["起始时间", "终止时间", "字幕"]
 
@@ -185,7 +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"]
@@ -231,7 +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)
@@ -260,7 +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:
@@ -279,8 +279,16 @@ class ProjectContext:
         # todo:现在是只用None判断是否是字幕，后续是否也需要用""来？
         for i in range(len(d["字幕"])):
             st_time_sec, ed_time_sec, subtitle, suggest, aside, speed = [d[x][i] for x in self.header]
+            print(">>>>>>>>>>>>>>load ed time")
+            print(st_time_sec)
+            print(ed_time_sec)
+            if ed_time_sec == None or ed_time_sec == "":
+                print(">>>>into load")
+                ed_time_sec = st_time_sec
             # 当前条目是字幕
             if d["字幕"][i] != None:
+                st_time_sec = reverse_time_to_seconds(str(st_time_sec))
+                ed_time_sec = reverse_time_to_seconds(str(ed_time_sec))
                 self.subtitle_list.append(Element(st_time_sec, ed_time_sec, subtitle, suggest, aside, speed))
                 self.all_elements.append(self.subtitle_list[-1])
             else:
@@ -299,12 +307,22 @@ class ProjectContext:
                 if d["终止时间"][i] is None:
                     # 如果是最后一条
                     if i == len(d["字幕"]) - 1:
-                        ed_time_sec = "360000" if self.duration == 0 else self.duration # todo 默认最大时长是100h
+                        print(1)
+                        # ed_time_sec = "360000" if self.duration == 0 else self.duration # todo 默认最大时长是100h
                     else:
                         ed_time_sec = "%.2f"%(float(d["起始时间"][i + 1]) - 0.01)
                 else:
                     ed_time_sec = d["终止时间"][i]
-
+                st_time_sec = reverse_time_to_seconds(str(st_time_sec))
+                ed_time_sec = reverse_time_to_seconds(str(ed_time_sec))
+                if aside == None:
+                    aside = ""
+                if suggest != None and suggest != "":
+                    arrays = suggest.split("/")
+                    if len(arrays) == 2:
+                        suggest =  str(len(aside)) + "/" + arrays[1]
+                    else:
+                        suggest =  str(len(aside))  + "/" + arrays[0]
                 self.aside_list.append(Element(st_time_sec, ed_time_sec, subtitle, suggest, aside, speed))
                 self.all_elements.append(self.aside_list[-1])
             # print("[load_excel_from_path] ", end='')
@@ -370,8 +388,9 @@ 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):
@@ -450,17 +469,31 @@ def write_to_sheet(path: str, sheet_name: str, valuelist: list):
         value = value_element.to_list()
         # 把None换成空串
         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.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] = 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]
+        if suggest != None and suggest != "":
+            arrays = suggest.split("/")
+            if len(arrays) == 2:
+                value[3] =  str(len(value[4]))  + "/" + arrays[1]
+            else:
+                value[3] =  str(len(value[4]))  + "/" + arrays[0]
+
         index = len(value)
         cur_row = sheet.max_row
         for j in range(0, index):
-            sheet.cell(row=cur_row + 1, column=j + 1, value=str(value[j]))
+            cell = sheet.cell(row=cur_row + 1, column=j + 1, value=str(value[j]))
             if value[j] == '' or '插入旁白' in str(value[j]):
-                sheet.cell(row=cur_row + 1, column=j + 1).fill = PatternFill(fill_type='solid', fgColor='ffff00')
+                cell = sheet.cell(row=cur_row + 1, column=j + 1).fill = PatternFill(fill_type='solid', fgColor='ffff00')
             if j == 4:
-                sheet.cell(row=cur_row + 1, column=j + 1).alignment = Alignment(wrapText=True)
+                cell = sheet.cell(row=cur_row + 1, column=j + 1).alignment = Alignment(wrapText=True)
+            cell.number_format = '@'
     workbook.save(path)
 
 

myvideoslider.py

@@ -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

ocr_metric.py

+import re
+import csv
+import argparse
+import pandas as pd
+import numpy as np
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
+from difflib import SequenceMatcher
+title = ['起始时间（转换后）', '终止时间（转换后）', '字幕']
+
+def init():
+    # 获取中文停用词列表
+    global stop_words
+    with open('chinese_stopwords.txt', 'r', encoding='utf-8') as file:
+        stop_words = set(line.strip() for line in file)
+
+# 将保存的时间戳转化为秒
+def change_to_second(time_str):
+    from datetime import datetime
+    time_obj = datetime.strptime(time_str, "%H:%M:%S,%f")
+    seconds = time_obj.hour * 3600 + time_obj.minute * 60 + \
+              time_obj.second + time_obj.microsecond / 1000000
+    return seconds
+
+# 计算字幕的相似度
+def calculate_similarity(str1, str2, method='cosine'):
+    if method == 'cosine':
+        tfidf_vectorizer = TfidfVectorizer()
+        tfidf_matrix = tfidf_vectorizer.fit_transform([str1, str2])
+        similarity_matrix = cosine_similarity(tfidf_matrix)
+        return similarity_matrix[0][1]
+    else :
+        return SequenceMatcher(None, str1, str2).ratio()
+
+# 计算两个时间戳的时间差（单位：秒）
+def calculate_time_difference(time1, time2):
+    return abs(time2 - time1)
+
+def calculate_weight(x, y):
+    # weight = e^(-alpha * time_diff)
+    # 相差1s的系数为0.9
+    alpha = 0.11
+    return 1 / (alpha * (x + y) + 1)
+
+# 检查句子中的每个单词是否都是停用词
+def is_all_stopwords(sentence):
+    return all(word in stop_words for word in sentence)
+
+### 如果其中有-符号，可能在用excel打开时自动添加=变成公式，读取的时候没问题
+def read_srt_to_csv(path_srt, path_output):
+    with open(path_srt, 'r', encoding='utf-8-sig') as f:
+        srt_content = f.read()  # str
+    # 使用正则表达式匹配时间码和字幕内容
+    pattern = re.compile(r'(\d+)\n([\d:,]+) --> ([\d:,]+)\n(.+?)(?=\n\d+\n|$)', re.DOTALL)
+    matches = pattern.findall(srt_content)
+
+    # 写入 csv 文件
+    with open(path_output, 'w', newline='', encoding='utf-8') as f:
+        csv_writer = csv.writer(f)
+        csv_writer.writerow(title)
+
+        for _, start, end, subtitle in matches:  # 都是str格式
+            subtitle = re.sub(r'\{[^}]*\}', '', subtitle)  # 将srt文件前的加粗等格式去掉
+            csv_writer.writerow([start, end, subtitle.strip()])
+
+def read_from_xlsx(path_xlsx='output.xlsx', path_output='deal.csv'):
+    data = pd.read_excel(path_xlsx)
+    with open(path_output, 'w', newline='', encoding='utf-8') as f:
+        csv_writer = csv.writer(f)
+        csv_writer.writerow(title)
+
+        for _, data1 in data.iterrows():
+            start, end, subtitle = data1[1], data1[3], data1[4]
+            if isinstance(subtitle, float) and np.isnan(subtitle):
+                continue
+            # 与srt文件格式同步
+            start = start.replace('.', ',')
+            end = end.replace('.', ',')
+            # print(start, end, subtitle,)
+            # print(type(start), type(end), type(subtitle))
+            csv_writer.writerow([start, end, subtitle.strip()])
+
+### 对于srt中的字幕计算相似性度。从ocr中找到时间戳满足<=time_t的字幕，
+### 然后计算字幕间的相似度，取一个最大的。字幕从start和end都匹配一遍
+# time_threshold设置阈值，用于判断时间差是否可接受
+def measure_score(path_srt, path_ocr, time_threshold=5.0, method='cosine'):
+    data_srt, data_ocr = [], []
+    with open(path_srt, 'r', encoding='utf-8') as file:
+        csv_reader = csv.reader(file)
+        data_srt = [i for i in csv_reader]
+        data_srt.pop(0)
+        for i in range(len(data_srt)):
+            data_srt[i][0] = change_to_second(data_srt[i][0])
+            data_srt[i][1] = change_to_second(data_srt[i][1])
+    with open(path_ocr, 'r', encoding='utf-8') as file:
+        csv_reader = csv.reader(file)
+        data_ocr = [i for i in csv_reader]
+        data_ocr.pop(0)
+        for i in range(len(data_ocr)):
+            data_ocr[i][0] = change_to_second(data_ocr[i][0])
+            data_ocr[i][1] = change_to_second(data_ocr[i][1])
+
+    # 计算相似度
+    total_similarity = 0.0
+    total_weight = 0.0
+
+    for sub in data_srt:
+        max_similarity = 0.0
+        # 去除srt中的停用词
+        if is_all_stopwords(sub[2]):
+            continue
+        for sub1 in data_ocr:
+            x, y = abs(sub[0] - sub1[0]), abs(sub[1] - sub1[1])
+            if min(x, y) <= time_threshold:
+                # print(sub[2], sub1[2])
+                score = calculate_similarity(sub[2], sub1[2], 'cosine')
+                max_similarity = max(max_similarity, score * calculate_weight(x, y))
+        total_similarity += max_similarity
+        total_weight += 1
+
+    # print(total_similarity, total_similarity / len(data_srt), total_similarity / total_weight)
+    return total_similarity / len(data_srt), total_similarity / total_weight
+
+if __name__ == '__main__':
+    init()
+    parser = argparse.ArgumentParser(description="benchmark")
+
+    # 添加命令行参数
+    parser.add_argument("--path_srt", required=True, type=str, help="Path of srt file, format is srt")
+    parser.add_argument("--path_ocr", required=True, type=str, help="Path of ocr file, format is xlsx")
+    parser.add_argument("--method", type=str, default='cosine', help="Select evaluation method")
+    parser.add_argument("--time_threshold", type=float,default=5.0, help="Allowable time frame")
+    args = parser.parse_args()
+
+    output_file_srt = 'deal_srt.csv'
+    output_file_ocr = 'deal_ocr.csv'
+    read_srt_to_csv(args.path_srt, output_file_srt)
+    read_from_xlsx(args.path_ocr, output_file_ocr)
+    score = measure_score(output_file_srt, output_file_ocr, args.time_threshold, args.method)
+    print(f'该评估算法得分: {score[1]:.5f}')
\ No newline at end of file

operation_dialog.py

@@ -260,9 +260,12 @@ class Operation_Dialog(QDialog, Ui_Dialog):
         else:  # 如果是旁白
             end_time = ""
             subtitle = ""
-            suggest = "插入旁白，推荐字数为" + suggest
+            # suggest = "插入旁白，推荐字数为" + suggest
+            # if self.comboBox_2.currentText() == "增加一行":
+            #     suggest = "插入旁白，推荐字数为0"
+            suggest = suggest
             if self.comboBox_2.currentText() == "增加一行":
-                suggest = "插入旁白，推荐字数为0"
+                suggest = "0/100"
         # 别忘复原
         self.buttonBox.setEnabled(False)
         self.zmpb_change_slot()

render.py

@@ -14,6 +14,7 @@ from PyQt5.QtWidgets import *;
 
 from management import RunThread
 from speech_synthesis import ffmpeg_path
+from ffmpeg_util import cg_wav
 
 class ExportProcessor(QWidget):
     show_warning_signal = pyqtSignal(str)
@@ -23,27 +24,30 @@ class ExportProcessor(QWidget):
         self.state = [0]
         self.threads = []
 
-    def export_slot(self, video_path, output_dir):
+    def export_slot(self, video_path, output_dir, pb_cg_rate):
         t = RunThread(funcName=self.start_export,
-                       args=(video_path, output_dir),
+                       args=(video_path, output_dir, pb_cg_rate),
                        name="export")
         t.setDaemon(True)
         self.threads.append(t)
 
         for t in self.threads:
-            t.start()
+            try:
+                t.start()
+            except Exception as e:
+                print(e)
         
         print("===子线程已经开启 in export===")
 
         self.export_callback_signal.emit(self.threads, self.state)
     
-    def start_export(self, video_path, output_dir):
-        mixed_audio_path = aggrevate_audios(video_path, output_dir, self.state)
+    def start_export(self, video_path, output_dir, pb_cg_rate):
+        mixed_audio_path = aggrevate_audios(video_path, output_dir, self.state, pb_cg_rate)
         export_video(video_path, mixed_audio_path, output_dir, self.state)
 
 
     # 生成一条无声的音频，然后把旁白音频逐个按照时间位置放进去，得到仅含旁白的音频和旁白+原声的音频
-def aggrevate_audios(video_path: str, output_dir: str, state=None):
+def aggrevate_audios(video_path: str, output_dir: str, state=None,pb_cg_rate = 1.00):
     # 这个模块最多只有80%的进度
     if state is None:
         state = [0]
@@ -56,6 +60,18 @@ def aggrevate_audios(video_path: str, output_dir: str, state=None):
 
     # 将生成的旁白音频放入空白音频中，并将原音频的对应位置音量降低为原来的30%
     files = os.listdir(output_dir)
+
+    if pb_cg_rate != 1.00:
+        audio_path = os.path.join(output_dir, os.path.basename(video_path).split('.')[0] + ".wav")
+        for i, f in enumerate(files):
+            fname = '.'.join(f.split('.')[:-1])
+            try:
+                if fname.find(".") != -1:
+                    cg_wav(audio_path, os.path.join(output_dir, f), pb_cg_rate)
+            except Exception as e:
+                print(e)
+                continue
+    
     for i, f in enumerate(files):
         fname = '.'.join(f.split('.')[:-1])
         try:

res/conf.ini

-{"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": "test\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

 {
-    "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,
-        "name": "test",
+        "name": "eagle",
         "language": "中文（普通话，简体）",
         "age_group": "年轻人",
         "gender": "女",

setting_dialog.py

@@ -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

@@ -23,11 +23,12 @@ 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
+from utils import reverse_time_to_seconds
 
 tmp_file = 'tmp.wav'
 adjusted_wav_path = "adjusted.wav"
@@ -81,8 +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
 
@@ -105,41 +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):
@@ -205,7 +210,7 @@ def get_narratage_text(sheet_content: dict) -> Tuple[list, list, list]:
     for i, text in enumerate(narratage):
         # 这里的speed是x.x倍速
         speed = float(speeds[i].split('(')[0])
-        if text is not None:
+        if text is not None and text != "":
             if text == '翻译':
                 narratage_text.append(subtitle[i])
             else:
@@ -236,8 +241,10 @@ def get_narratage_text(sheet_content: dict) -> Tuple[list, list, list]:
                 #     narratage_start_time.append(cur_start)
                 #     narratage_end_time.append(cur_end)
                 #     cur_start = cur_start + (len(x) / normal_speed + normal_interval) / speed
-            narratage_start_time.append(float(start_time[i]))
-            narratage_end_time.append(float(end_time[i])) if end_time[i] is not None else narratage_end_time.append(-1)
+            # narratage_start_time.append(float(start_time[i]))
+            # narratage_end_time.append(float(end_time[i])) if end_time[i] is not None else narratage_end_time.append(-1)
+            narratage_start_time.append(float(reverse_time_to_seconds(start_time[i])))
+            narratage_end_time.append(float(reverse_time_to_seconds(end_time[i]))) if end_time[i] is not None and end_time[i] != "" else narratage_end_time.append(-1)
             narratage_speed.append(speed)
     return narratage_text, narratage_start_time, narratage_end_time, narratage_speed
 
@@ -251,13 +258,16 @@ def second_to_str(seconds: float) -> str:
     Returns:
         str: “时:分:秒”格式的时间字符串
     """
-    seconds = float(seconds)
-    hour = int(seconds / 3600)
-    minute = int((seconds - hour * 3600) / 60)
-    second = int(seconds - hour * 3600 - minute * 60)
-    ms = int((seconds - second - minute * 60 - hour * 3600) * 1000)
-    time_str = "%02d:%02d:%02d,%03d" % (hour, minute, second, ms)
-    return time_str
+    try:
+        seconds = float(seconds)
+        hour = int(seconds / 3600)
+        minute = int((seconds - hour * 3600) / 60)
+        second = int(seconds - hour * 3600 - minute * 60)
+        ms = int((seconds - second - minute * 60 - hour * 3600) * 1000)
+        time_str = "%02d:%02d:%02d,%03d" % (hour, minute, second, ms)
+        return time_str
+    except:
+        return str(seconds)
 
 
 def export_caption(sheet_content: dict, caption_file: str):

start.py

@@ -56,19 +56,34 @@ def change_project_path(path):
 if __name__ == '__main__':
     try:
         # subprocess.call(['deploy.bat'])
-        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 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)
+        #     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
+        QCoreApplication.setAttribute(Qt.AA_EnableHighDpiScaling)
+        QCoreApplication.setAttribute(Qt.AA_UseHighDpiPixmaps)
+        currentExitCode = MainWindow.EXIT_CODE_REBOOT
         while currentExitCode == MainWindow.EXIT_CODE_REBOOT:
             app = QApplication(sys.argv)
             app.setWindowIcon(QIcon("./res/images/eagle_2.ico"))
@@ -82,6 +97,8 @@ if __name__ == '__main__':
             apply_stylesheet(app, theme='dark_amber.xml')
             # mainWindow.show()
             mainWindow.showMaximized()
+            if not os.path.exists("C:\LavFilters") and not os.path.exists("C:\Program Files (x86)\LAV Filters"):
+                mainWindow.import_excel_dialog.show_with_msg("视频解码器未正常安装")
             currentExitCode = app.exec_()
             app = None
     except Exception as e:

start.spec

@@ -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 = []
@@ -20,8 +20,8 @@ a = Analysis(
     ['start.py'],
     pathex=[],
     binaries=[],
-    datas=[('res','res')] + pkgPaths,
-    hiddenimports=[],
+    datas=[('vits_chinese','vits_chinese')] + [('res','res'),('./deploy.bat','.'),('./ding_notify.conf','.'),('LAVFilters64','LAVFilters64'),('LAVFilters32','LAVFilters32')] + pkgPaths,
+    hiddenimports=['filecmp'],
     hookspath=[],
     hooksconfig={},
     runtime_hooks=[],
@@ -38,7 +38,7 @@ exe = EXE(
     a.scripts,
     [],
     exclude_binaries=True,
-    name='start',
+    name='无障碍电影制作系统',
     debug=False,
     bootloader_ignore_signals=False,
     strip=False,
@@ -49,6 +49,7 @@ exe = EXE(
     target_arch=None,
     codesign_identity=None,
     entitlements_file=None,
+    icon=['res\\images\\eagle_2.ico'],
 )
 coll = COLLECT(
     exe,
@@ -58,5 +59,5 @@ coll = COLLECT(
     strip=False,
     upx=True,
     upx_exclude=[],
-    name='start',
+    name='无障碍电影制作系统',
 )

utils.py

@@ -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
 
@@ -47,13 +70,17 @@ 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
@@ -64,13 +91,75 @@ def transfer_second_to_time(sec: str) -> str:
     Returns:
         str: "hh:mm:ss.xxx"格式的时间字符串
     """
-    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)
-    return time
+    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
+
+    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 reverse_time_to_seconds(time_str: str) -> str:
+    """将"hh:mm:ss.xxx"格式的时间字符串转换为秒数
+
+    Args:
+        time_str (str): 时间字符串，格式为"hh:mm:ss.xxx"
+
+    Returns:
+        float: 对应的秒数
+    """
+    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
+        total_seconds = hour * 3600 + minutes * 60 + seconds
+        print(str(total_seconds))
+        return str(total_seconds)
+    except Exception as e:
+        print(">>>>>>>>>>reverse time err" + time_str)
+        print(e)
+        return time_str
 
 def replace_path_suffix(path: str, new_suffix: str) -> str:
     """替换文件路径后缀
@@ -93,7 +182,7 @@ def check_sheet_content(book_path: str) -> bool:
         bool: True(用户输入的表符合预期) or False(用户输入的表格式不正确，很可能输入错误)
     """
     sheet_heads = get_sheetHead(book_path)
-    need_heads = ['起始时间', '终止时间', '字幕', '建议', '解说脚本']
+    need_heads = ['起始时间', '终止时间', '字幕', '建议旁白字数', '解说脚本']
 
     if len(sheet_heads) == 0:
         return False
@@ -150,7 +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)