10 Commits
5 changed files with 107 additions and 50 deletions
+6 -8
View File
@@ -1,8 +1,6 @@
- Added voice mixing, allowing multiple voices to be combined into a single “Mixed Voice”, a feature mentioned by @PulsarFTW in #1. Special thanks to @jborza for making this possible through his contributions in #5. - Merge pull request [#7](https://github.com/denizsafak/abogen/pull/7) by [@jborza](https://github.com/jborza) that improves voice preview and documentation.
- Added profile system to voice mixer, allowing users to create and manage multiple voice profiles. - Fixed the issue when a voice is selected, the voice mixer tries to pre-select that voice and ignores existing profiles.
- Improvements in the voice mixer, mostly for organizing controls and enhancing user experience. - Fixed the error while renaming the default "New profile" in the voice mixer.
- Added icons for flags and genders in the GUI, making it easier to identify different options. - Fixed subtitle_combo enabling/disabling when a voice in the voice mixer is selected.
- Improved the content and chapter extraction process for EPUB files, ensuring better handling of various structures. - Prevented using special characters in the profile name to avoid conflicts.
- Switched to platformdirs for determining the correct desktop path, instead of using old methods. - Improved invalid profile handling in the voice mixer.
- Fixed preview voices was not using GPU acceleration, which was causing performance issues.
- Improvements in code and documentation.
+1
View File
@@ -160,6 +160,7 @@ If you'd like to modify the code and contribute to development, you can [downloa
```bash ```bash
# Go to the directory where you extracted the repository and run: # Go to the directory where you extracted the repository and run:
pip install -e . # Installs the package in editable mode pip install -e . # Installs the package in editable mode
pip install build # Install the build package
python -m build # Builds the package in dist folder (optional) python -m build # Builds the package in dist folder (optional)
abogen # Opens the GUI abogen # Opens the GUI
``` ```
+1 -1
View File
@@ -1 +1 @@
1.0.3 1.0.4
+26 -7
View File
@@ -1480,8 +1480,11 @@ class abogen(QWidget):
lang = self.selected_voice[0] lang = self.selected_voice[0]
voice = self.selected_voice voice = self.selected_voice
# use same gpu/cpu logic as in conversion
gpu_msg, gpu_ok = get_gpu_acceleration(self.use_gpu)
self.preview_thread = VoicePreviewThread( self.preview_thread = VoicePreviewThread(
np_module, kpipeline_class, lang, voice, speed, self.use_gpu np_module, kpipeline_class, lang, voice, speed, gpu_ok
) )
self.preview_thread.finished.connect(self._play_preview_audio) self.preview_thread.finished.connect(self._play_preview_audio)
self.preview_thread.error.connect(self._preview_error) self.preview_thread.error.connect(self._preview_error)
@@ -1545,7 +1548,12 @@ class abogen(QWidget):
def _preview_cleanup(self): def _preview_cleanup(self):
self.preview_playing = False self.preview_playing = False
self.btn_preview.setIcon(self.play_icon) self.loading_movie.stop()
try:
self.loading_movie.frameChanged.disconnect()
except Exception:
pass # Ignore error if not connected
self.btn_preview.setIcon(self.play_icon)
self.btn_preview.setToolTip("Preview selected voice") self.btn_preview.setToolTip("Preview selected voice")
self.btn_preview.setEnabled(True) self.btn_preview.setEnabled(True)
self.voice_combo.setEnabled(True) self.voice_combo.setEnabled(True)
@@ -1820,25 +1828,36 @@ class abogen(QWidget):
save_config(self.config) save_config(self.config)
def show_voice_formula_dialog(self): def show_voice_formula_dialog(self):
# If a profile is selected in combo, open mixer with that profile from voice_profiles import load_profiles
profiles = load_profiles()
initial_state = None initial_state = None
selected_profile = self.selected_profile_name selected_profile = self.selected_profile_name
if selected_profile: if selected_profile:
entry = load_profiles().get(selected_profile, {}) entry = profiles.get(selected_profile, {})
# support new dict format
if isinstance(entry, dict): if isinstance(entry, dict):
initial_state = entry.get("voices", []) initial_state = entry.get("voices", [])
else: else:
initial_state = entry initial_state = entry
elif self.mixed_voice_state is not None: elif self.mixed_voice_state is not None:
initial_state = self.mixed_voice_state initial_state = self.mixed_voice_state
elif self.selected_voice:
# If a single voice is selected, default to first profile if available
if profiles:
first_profile = next(iter(profiles))
entry = profiles[first_profile]
selected_profile = first_profile
if isinstance(entry, dict):
initial_state = entry.get("voices", [])
else:
initial_state = entry
else:
initial_state = []
else: else:
initial_state = [(self.selected_voice, 1.0)] if self.selected_voice else [] initial_state = []
dialog = VoiceFormulaDialog( dialog = VoiceFormulaDialog(
self, initial_state=initial_state, selected_profile=selected_profile self, initial_state=initial_state, selected_profile=selected_profile
) )
if dialog.exec_() == QDialog.Accepted: if dialog.exec_() == QDialog.Accepted:
# After OK, select the profile in combo and update config
if dialog.current_profile: if dialog.current_profile:
self.selected_profile_name = dialog.current_profile self.selected_profile_name = dialog.current_profile
self.config["selected_profile_name"] = dialog.current_profile self.config["selected_profile_name"] = dialog.current_profile
+73 -34
View File
@@ -475,7 +475,6 @@ class VoiceFormulaDialog(QDialog):
self.add_voices(initial_state or []) self.add_voices(initial_state or [])
self.update_weighted_sums() self.update_weighted_sums()
self.update_subtitle_combo_enabled()
# assemble splitter # assemble splitter
splitter.addWidget(profile_widget) splitter.addWidget(profile_widget)
@@ -496,8 +495,6 @@ class VoiceFormulaDialog(QDialog):
for vm in self.voice_mixers: for vm in self.voice_mixers:
vm.spin_box.valueChanged.connect(self.mark_profile_modified) vm.spin_box.valueChanged.connect(self.mark_profile_modified)
vm.checkbox.stateChanged.connect(lambda *_: self.mark_profile_modified()) vm.checkbox.stateChanged.connect(lambda *_: self.mark_profile_modified())
vm.spin_box.valueChanged.connect(self.update_subtitle_combo_enabled)
vm.checkbox.stateChanged.connect(self.update_subtitle_combo_enabled)
def keyPressEvent(self, event): def keyPressEvent(self, event):
# Bind Delete key to delete_profile when a profile is selected # Bind Delete key to delete_profile when a profile is selected
@@ -682,8 +679,6 @@ class VoiceFormulaDialog(QDialog):
voice_mixer.checkbox.stateChanged.connect( voice_mixer.checkbox.stateChanged.connect(
lambda *_: self.mark_profile_modified() lambda *_: self.mark_profile_modified()
) )
voice_mixer.spin_box.valueChanged.connect(self.update_subtitle_combo_enabled)
voice_mixer.checkbox.stateChanged.connect(self.update_subtitle_combo_enabled)
return voice_mixer return voice_mixer
def handle_voice_checkbox(self, voice_mixer, state): def handle_voice_checkbox(self, voice_mixer, state):
@@ -743,20 +738,6 @@ class VoiceFormulaDialog(QDialog):
self.error_label.show() self.error_label.show()
self.weighted_sums_container.hide() self.weighted_sums_container.hide()
def update_subtitle_combo_enabled(self):
# Only enable subtitle_combo if at least one selected voice is from supported languages
selected_langs = set()
for vm in self.voice_mixers:
if vm.checkbox.isChecked() and vm.spin_box.value() > 0:
lang_code = vm.voice_name[0]
selected_langs.add(lang_code)
enable = any(
lang in SUPPORTED_LANGUAGES_FOR_SUBTITLE_GENERATION
for lang in selected_langs
)
if self.subtitle_combo:
self.subtitle_combo.setEnabled(enable)
def disable_voice_by_name(self, voice_name): def disable_voice_by_name(self, voice_name):
for mixer in self.voice_mixers: for mixer in self.voice_mixers:
if mixer.voice_name == voice_name: if mixer.voice_name == voice_name:
@@ -819,7 +800,6 @@ class VoiceFormulaDialog(QDialog):
# sync enabled state # sync enabled state
vm.toggle_inputs() vm.toggle_inputs()
self.update_weighted_sums() self.update_weighted_sums()
self.update_subtitle_combo_enabled()
def save_profile_by_name(self, name): def save_profile_by_name(self, name):
profiles = load_profiles() profiles = load_profiles()
@@ -846,10 +826,10 @@ class VoiceFormulaDialog(QDialog):
self.update_weighted_sums() self.update_weighted_sums()
def _handle_zero_weight_profiles(self): def _handle_zero_weight_profiles(self):
if self.profile_list.count() <= 1: profiles = load_profiles()
if len(profiles) < 1:
return False return False
zero = [] zero = []
profiles = load_profiles()
for i in range(self.profile_list.count()): for i in range(self.profile_list.count()):
item = self.profile_list.item(i) item = self.profile_list.item(i)
name = item.text().lstrip("*") name = item.text().lstrip("*")
@@ -868,7 +848,6 @@ class VoiceFormulaDialog(QDialog):
if not zero: if not zero:
return False return False
msg = f"{len(zero)} invalid profile(s) with no voices selected or their total weights are 0. They will be ignored and deleted. Do you want to delete?" msg = f"{len(zero)} invalid profile(s) with no voices selected or their total weights are 0. They will be ignored and deleted. Do you want to delete?"
# Use Delete instead of Ignore
reply = QMessageBox.question( reply = QMessageBox.question(
self, self,
"Invalid Profiles", "Invalid Profiles",
@@ -963,8 +942,17 @@ class VoiceFormulaDialog(QDialog):
self.update_profile_list_colors() self.update_profile_list_colors()
def new_profile(self): def new_profile(self):
name, ok = QInputDialog.getText(self, "New Profile", "Enter profile name:") import re
if ok and name: while True:
name, ok = QInputDialog.getText(self, "New Profile", "Enter profile name:")
if not ok or not name:
break
name = name.strip() # Remove leading/trailing spaces
if not name:
continue
if not re.match(r'^[\w\- ]+$', name):
QMessageBox.warning(self, "Invalid Name", "Profile name can only contain letters, numbers, spaces, underscores, and hyphens.")
continue
profiles = load_profiles() profiles = load_profiles()
# Remove 'New profile' placeholder if not persisted in JSON # Remove 'New profile' placeholder if not persisted in JSON
if ( if (
@@ -977,7 +965,7 @@ class VoiceFormulaDialog(QDialog):
self._profile_dirty.pop("New profile", None) self._profile_dirty.pop("New profile", None)
if name in profiles: if name in profiles:
QMessageBox.warning(self, "Duplicate Name", "Profile already exists.") QMessageBox.warning(self, "Duplicate Name", "Profile already exists.")
return continue
profiles[name] = { profiles[name] = {
"voices": [], "voices": [],
"language": self.language_combo.currentData(), "language": self.language_combo.currentData(),
@@ -996,6 +984,7 @@ class VoiceFormulaDialog(QDialog):
parent = self.parent() parent = self.parent()
if hasattr(parent, "populate_profiles_in_voice_combo"): if hasattr(parent, "populate_profiles_in_voice_combo"):
parent.populate_profiles_in_voice_combo() parent.populate_profiles_in_voice_combo()
break
self.update_profile_save_buttons() self.update_profile_save_buttons()
self.update_profile_list_colors() self.update_profile_list_colors()
self.update_weighted_sums() self.update_weighted_sums()
@@ -1182,21 +1171,64 @@ class VoiceFormulaDialog(QDialog):
) )
def rename_profile(self, item): def rename_profile(self, item):
name = item.text().lstrip("*")
# block if profile has unsaved changes and it's not a virtual New profile
if self._profile_dirty.get(name, False) and not (self._virtual_new_profile and name == "New profile"):
QMessageBox.warning(
self, "Unsaved Changes", "Please save the profile before renaming."
)
return
old = item.text().lstrip("*") old = item.text().lstrip("*")
new, ok = QInputDialog.getText( import re
self, "Rename Profile", f"Profile name:", text=old while True:
) new, ok = QInputDialog.getText(
if ok and new and new != old: self, "Rename Profile", f"Profile name:", text=old
)
if not ok or not new or new == old:
break
new = new.strip() # Remove leading/trailing spaces
if not new:
continue
if not re.match(r'^[\w\- ]+$', new):
QMessageBox.warning(self, "Invalid Name", "Profile name can only contain letters, numbers, spaces, underscores, and hyphens.")
continue
profiles = load_profiles() profiles = load_profiles()
if new in profiles: if new in profiles:
QMessageBox.warning(self, "Duplicate Name", "Profile already exists.") QMessageBox.warning(self, "Duplicate Name", "Profile already exists.")
return continue
profiles[new] = profiles.pop(old)
save_profiles(profiles) # Special case for renaming the virtual "New profile"
item.setText(new) if self._virtual_new_profile and name == "New profile":
# Create the profile with the new name
profiles[new] = {
"voices": self.get_selected_voices(),
"language": self.language_combo.currentData(),
}
save_profiles(profiles)
# Update tracking properties
self._virtual_new_profile = False
self._profile_dirty.pop("New profile", None)
self._profile_dirty[new] = False
# Update the current profile name
self.current_profile = new
item.setText(new)
else:
# Standard renaming for regular profiles
profiles[new] = profiles.pop(old)
save_profiles(profiles)
item.setText(new)
# Update the current profile name if it was renamed
if self.current_profile == old:
self.current_profile = new
parent = self.parent() parent = self.parent()
if hasattr(parent, "populate_profiles_in_voice_combo"): if hasattr(parent, "populate_profiles_in_voice_combo"):
parent.populate_profiles_in_voice_combo() parent.populate_profiles_in_voice_combo()
break
self.update_profile_save_buttons() self.update_profile_save_buttons()
self.update_profile_list_colors() self.update_profile_list_colors()
@@ -1227,6 +1259,13 @@ class VoiceFormulaDialog(QDialog):
self.update_profile_list_colors() self.update_profile_list_colors()
def duplicate_profile(self, item): def duplicate_profile(self, item):
name = item.text().lstrip("*")
# block duplicating if profile has unsaved changes
if self._profile_dirty.get(name, False):
QMessageBox.warning(
self, "Unsaved Changes", "Please save the profile before duplicating."
)
return
src = item.text().lstrip("*") src = item.text().lstrip("*")
profiles = load_profiles() profiles = load_profiles()
base = f"{src}_duplicate" base = f"{src}_duplicate"