0.8.1

CheshireCC · Sep 16, 2024 · 7ef7369 · 7ef7369
1 parent 66b7450
commit 7ef7369
Show file tree

Hide file tree

Showing 15 changed files with 36 additions and 31 deletions.
diff --git a/config/config.json b/config/config.json
@@ -1,6 +1,6 @@
 {
     "QFluentWidgets": {
-        "ThemeColor": "#ff152cff",
+        "ThemeColor": "#ffe64ecf",
         "ThemeMode": "Light"
     }
 }
diff --git a/fasterWhisperGUIConfig.json b/fasterWhisperGUIConfig.json
@@ -8,9 +8,9 @@
     "model_param": {
         "localModel": true,
         "onlineModel": false,
-        "model_path": "F:/WhisperModels/faster-whisper/large-v3-float32",
+        "model_path": "F:/WhisperModels/faster-whisper/whisper-large-v2-ct2-32",
         "modelName": 11,
-        "use_v3_model": true,
+        "use_v3_model": false,
         "device": 1,
         "deviceIndex": "0",
         "preciese": 5,
@@ -21,27 +21,26 @@
     },
     "vad_param": {
         "use_VAD": true,
-        "threshold": 0.5,
+        "threshold": 0.2,
         "minSpeechDuration": "250",
         "minSilenceDuration": "2000",
         "maxSpeechDuration": "inf",
-        "windowSize": 1,
         "speechPad": "400"
     },
     "setting": {
         "saveConfig": true,
         "autoLoadModel": false,
         "language": 2,
         "huggingface_user_token": "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",
-        "autoGoToOutputPage": 0,
+        "autoGoToOutputPage": 2,
         "autoClearTempFiles": true,
-        "themeColor": "#152cff"
+        "themeColor": "#e64ecf"
     },
     "Transcription_param": {
         "aggregate_contents": true,
-        "language": 2,
+        "language": -1,
         "task": false,
-        "beam_size": "5",
+        "beam_size": "1",
         "best_of": "5",
         "patience": "1.0",
         "length_penalty": "1.0",
@@ -56,7 +55,7 @@
         "suppress_tokens": "-1",
         "without_timestamps": false,
         "max_initial_timestamp": "1.0",
-        "word_timestamps": false,
+        "word_timestamps": true,
         "prepend_punctuations": "\"'“¿([{-",
         "append_punctuations": "\"'.。,，!！?？:：”)]}、",
         "repetition_penalty": "1.0",

diff --git a/faster_whisper_GUI/UI_MainWindows.py b/faster_whisper_GUI/UI_MainWindows.py
@@ -244,7 +244,7 @@ def setupUI(self):
         # 设置窗体中心控件
         self.setCentralWidget(self.mainWindowsWidget)
 
-        # TODO: 创建一个空对象 用于改善布局顶部
+        # 创建一个空对象 用于改善布局顶部
         self.spacer_main = QSpacerItem(0,25)
         self.vBoxLayout.addItem(self.spacer_main)
 

diff --git a/faster_whisper_GUI/aboutPageNavigationInterface.py b/faster_whisper_GUI/aboutPageNavigationInterface.py
@@ -8,6 +8,7 @@
                                 QGraphicsView, 
                                 QGraphicsScene
                             )
+
 from qfluentwidgets import (
                             DisplayLabel, 
                             ScrollArea, 

diff --git a/faster_whisper_GUI/mainWindows.py b/faster_whisper_GUI/mainWindows.py
@@ -31,7 +31,7 @@
                         )
 
 from faster_whisper.transcribe import TranscriptionInfo
-from faster_whisper import Word
+from faster_whisper.transcribe import Word
 
 import torch
 
@@ -866,15 +866,15 @@ def getVADparam(self) -> dict:
         min_speech_duration_ms = int(self.page_VAD.LineEdit_VAD_param_min_speech_duration_ms.text().replace(" ", ""))
         max_speech_duration_s = float(self.page_VAD.LineEdit_VAD_param_max_speech_duration_s.text().replace(" ", ""))
         min_silence_duration_ms = int(self.page_VAD.LineEdit_VAD_param_min_silence_duration_ms.text().replace(" ", ""))
-        window_size_samples = int(self.page_VAD.combox_VAD_param_window_size_samples.currentText())
+        # window_size_samples = int(self.page_VAD.combox_VAD_param_window_size_samples.currentText())
         speech_pad_ms = int(self.page_VAD.LineEdit_VAD_param_speech_pad_ms.text().replace(" ", ""))
 
         VAD_param["param"] = VADParameters()
         VAD_param["param"]["threshold"] = threshold
         VAD_param["param"]["min_speech_duration_ms"] = min_speech_duration_ms
         VAD_param["param"]["max_speech_duration_s"] = max_speech_duration_s
         VAD_param["param"]["min_silence_duration_ms"] = min_silence_duration_ms
-        VAD_param["param"]["window_size_samples"] = window_size_samples
+        # VAD_param["param"]["window_size_samples"] = window_size_samples
         VAD_param["param"]["speech_pad_ms"] = speech_pad_ms
 
         return VAD_param
@@ -1659,7 +1659,7 @@ def closeEvent(self, event) -> None:
             outputWithDateTime("Exit")
 
             if self.page_setting.switchButton_saveConfig.isChecked():
-                self.saveConfig(config_file_name=r'./fasterWhisperGUIConfig.json')
+                self.saveConfig(config_file_name=os.path.abspath(r'./fasterWhisperGUIConfig.json'))
 
             if self.page_setting.switchButton_autoClearTempFiles.isChecked():
                 try:

diff --git a/faster_whisper_GUI/seg_ment.py b/faster_whisper_GUI/seg_ment.py
@@ -1,6 +1,6 @@
 # coding:utf-8
 
-from faster_whisper import  (Segment, Word)
+from faster_whisper.transcribe import Segment, Word
 from typing import List
 
 class segment_Transcribe():

diff --git a/faster_whisper_GUI/subtitleFileRead.py b/faster_whisper_GUI/subtitleFileRead.py
@@ -4,7 +4,7 @@
 from typing import List
 from .seg_ment import segment_Transcribe
 import json
-from faster_whisper import Word
+from faster_whisper.transcribe import Word
 
 def readJSONFileToSegments(file:str, file_code = "utf8") -> List[segment_Transcribe]:
 

diff --git a/faster_whisper_GUI/tableModel_segments_path_info.py b/faster_whisper_GUI/tableModel_segments_path_info.py
@@ -5,7 +5,7 @@
 from typing import List
 from PySide6.QtGui import QColor
 from PySide6.QtWidgets import QApplication
-from faster_whisper import  Word
+from faster_whisper.transcribe import Word
 from qfluentwidgets import MessageBox, isDarkTheme
 
 from .seg_ment import segment_Transcribe

diff --git a/faster_whisper_GUI/transcribe.py b/faster_whisper_GUI/transcribe.py
@@ -12,10 +12,8 @@
 import json
 import hashlib
 
-from faster_whisper import (
-                            WhisperModel
-                            , TranscriptionInfo
-                        )
+from faster_whisper import WhisperModel
+from faster_whisper.transcribe import TranscriptionInfo
 
 import webvtt
 from PySide6.QtCore import (QThread, Signal, QDateTime)

diff --git a/faster_whisper_GUI/vadPageNavigationInterface.py b/faster_whisper_GUI/vadPageNavigationInterface.py
@@ -73,6 +73,10 @@ def setupUI(self):
                                                             self.__tr("语音概率阈值。 Silero VAD为每个音频块输出语音概率, 概率高于此值的认为是语音。\n最好对每个数据集单独调整此参数, 但“懒散”的 0.5 对大多数数据集来说都非常好。"),
                                                             self.doubleSpin_VAD_param_threshold
                                                         )
+
+
+        # self.VAD_param_threshold_param_widget.mainHLayout.setStretch(2,5)
+
 
         self.GridLayout_VAD_param.addWidget(self.VAD_param_threshold_param_widget, 0, 0)
 
@@ -113,12 +117,14 @@ def setupUI(self):
 
         combox_VAD_param_window_size_samples = ComboBox()
         combox_VAD_param_window_size_samples.addItems(["512", "1024", "1536"])
-        combox_VAD_param_window_size_samples.setCurrentIndex(1)
+        combox_VAD_param_window_size_samples.setCurrentIndex(0)
         combox_VAD_param_window_size_samples.setToolTip(self.__tr("警告!\nSilero VAD模型使用16000采样率训练得到512,1024,1536样本。其他值可能会影响模型性能!"))
+        combox_VAD_param_window_size_samples.setEnabled(False)
+
         self.combox_VAD_param_window_size_samples = combox_VAD_param_window_size_samples
 
         self.VAD_param_window_size_samples_param_widget = ParamWidget(self.__tr("采样窗口大小"),
-                                                                    self.__tr("指定大小的音频块被馈送到silero VAD模型。"),
+                                                                    self.__tr("指定大小的音频块被馈送到silero VAD模型。Silero V5 已经将该参数固定为 512 。"),
                                                                     self.combox_VAD_param_window_size_samples
                                                                 )
 
@@ -169,7 +175,7 @@ def getParam(self):
         param["minSpeechDuration"] = self.LineEdit_VAD_param_min_speech_duration_ms.text().strip()
         param["minSilenceDuration"] = self.LineEdit_VAD_param_min_silence_duration_ms.text().strip()
         param["maxSpeechDuration"] = self.LineEdit_VAD_param_max_speech_duration_s.text().strip()
-        param["windowSize"] = self.combox_VAD_param_window_size_samples.currentIndex()
+        # param["windowSize"] = self.combox_VAD_param_window_size_samples.currentIndex()
         param["speechPad"] = self.LineEdit_VAD_param_speech_pad_ms.text().strip()
 
         return param
@@ -181,6 +187,6 @@ def setParam(self, param:dict):
         self.LineEdit_VAD_param_min_speech_duration_ms.setText(param["minSpeechDuration"])
         self.LineEdit_VAD_param_min_silence_duration_ms.setText(param["minSilenceDuration"])
         self.LineEdit_VAD_param_max_speech_duration_s.setText(param["maxSpeechDuration"])
-        self.combox_VAD_param_window_size_samples.setCurrentIndex(param["windowSize"])
+        # self.combox_VAD_param_window_size_samples.setCurrentIndex(param["windowSize"])
         self.LineEdit_VAD_param_speech_pad_ms.setText(param["speechPad"])
 
diff --git a/faster_whisper_GUI/version.py b/faster_whisper_GUI/version.py
@@ -1,6 +1,6 @@
 # coding:utf-8
 
-__version__ = "0.8.0"
-__FasterWhisper_version__ = "1.0.2"
+__version__ = "0.8.1"
+__FasterWhisper_version__ = "1.0.3"
 __WhisperX_version__ = "3.1.1"
 __Demucs_version__ = "v4.0"
diff --git a/faster_whisper_GUI/whisper_x.py b/faster_whisper_GUI/whisper_x.py
@@ -73,6 +73,7 @@ def run(self):
                                                                                                 ,model_dir=r"./cache"
                                                                                                 ,cache_dir=r"./cache"
                                                                                             )
+
                     print("start alignment...")
                     self.setStateTool(text="start alignment...",status=False)
 

diff --git a/resource/_rc/rc_Image.py b/resource/_rc/rc_Image.py
@@ -1,6 +1,6 @@
 # Resource object code (Python 3)
 # Created by: object code
-# Created by: The Resource Compiler for Qt version 6.6.1
+# Created by: The Resource Compiler for Qt version 6.7.2
 # WARNING! All changes made in this file will be lost!
 
 from PySide6 import QtCore

diff --git a/resource/_rc/rc_Translater.py b/resource/_rc/rc_Translater.py
@@ -1,6 +1,6 @@
 # Resource object code (Python 3)
 # Created by: object code
-# Created by: The Resource Compiler for Qt version 6.6.1
+# Created by: The Resource Compiler for Qt version 6.7.2
 # WARNING! All changes made in this file will be lost!
 
 from PySide6 import QtCore

diff --git a/resource/_rc/rc_qss.py b/resource/_rc/rc_qss.py
@@ -1,6 +1,6 @@
 # Resource object code (Python 3)
 # Created by: object code
-# Created by: The Resource Compiler for Qt version 6.6.1
+# Created by: The Resource Compiler for Qt version 6.7.2
 # WARNING! All changes made in this file will be lost!
 
 from PySide6 import QtCore