Dymas/abogen
1
0
Fork 0
mirror of https://github.com/denizsafak/abogen.git synced 2026-07-18 13:40:27 +02:00
Code Issues Packages Projects Releases Wiki Activity

Restyle using black, improve m4b process handling

Browse Source
This commit is contained in:
Deniz Şafak
2025-05-04 11:39:54 +03:00
parent 412fb3bc7e
commit 146fcb3c1f
5 changed files with 104 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CHANGELOG.md
+1 -3
View File
@@ -1,3 +1 @@
- Added new output format: `m4b`, enabling chapter metadata in audiobooks. Special thanks to @jborza for implementing this feature in PR #10.
- Better approach for determining the correct configuration folder for Linux and MacOS, using platformdirs. (Fixes Docker issue #12)
- Improvements in documentation and code.
- Fixed m4b chapter generation opens CMD window in Windows.
abogen/conversion.py
+60 -15
View File
@@ -13,6 +13,10 @@ from constants import PROGRAM_NAME, LANGUAGE_DESCRIPTIONS, SAMPLE_VOICE_TEXTS
from voice_formulas import get_new_voice
import static_ffmpeg
import threading # for efficient waiting
import subprocess
import sys
import platform
def get_sample_voice_text(lang_code):
return SAMPLE_VOICE_TEXTS.get(lang_code, SAMPLE_VOICE_TEXTS["a"])
@@ -610,13 +614,16 @@ class ConversionThread(QThread):
# If there is only one chapter, skip adding chapters and just return the wav file path
if not chapters_time or len(chapters_time) <= 1:
self.log_updated.emit(
("File contains only one chapter or no chapters were detected. The audio will be saved as a standard .wav file instead.\n", "red")
(
"File contains only one chapter or no chapters were detected. The audio will be saved as a standard .wav file instead.\n",
"red",
)
)
return out_path
# generate chapters.txt in the same folder as the output file
chapters_info_path = os.path.splitext(out_path)[0] + "_chapters.txt"
with open(chapters_info_path, "w", encoding="utf-8") as f:
f.write(';FFMETADATA1\n') # required header for ffmpeg
f.write(";FFMETADATA1\n") # required header for ffmpeg
for chapter in chapters_time:
f.write(f"[CHAPTER]\n")
f.write(f"TIMEBASE=1/10\n")
@@ -627,29 +634,67 @@ class ConversionThread(QThread):
# call ffmpeg to merge the chapters into the output file
# ffmpeg installed on first call to add_paths()
static_ffmpeg.add_paths()
import subprocess
out_path_m4b = os.path.splitext(out_path)[0] + ".m4b"
# ffmpeg -i input.m4b -i ch.txt -map 0:a -map_chapters 1 output.m4b
ffmpeg_cmd = [
"ffmpeg",
"-i", out_path,
"-i", chapters_info_path,
"-map", "0:a",
"-map_chapters", "1",
out_path_m4b
"-i",
out_path,
"-i",
chapters_info_path,
"-map",
"0:a",
"-map_chapters",
"1",
out_path_m4b,
]
self.log_updated.emit("Adding chapters to the audio file...\n")
result = subprocess.run(ffmpeg_cmd, capture_output=True, text=True)
# Define kwargs for Popen
default_encoding = sys.getdefaultencoding() # Get default encoding
kwargs = {
"stdout": subprocess.PIPE,
"stderr": subprocess.PIPE, # Capture stderr separately
"universal_newlines": True,
"encoding": default_encoding,
"errors": "replace",
}
# Add Windows-specific settings
if platform.system() == "Windows":
startupinfo = subprocess.STARTUPINFO()
startupinfo.dwFlags |= subprocess.STARTF_USESHOWWINDOW
startupinfo.wShowWindow = subprocess.SW_HIDE
kwargs.update(
{
"startupinfo": startupinfo,
"creationflags": subprocess.CREATE_NO_WINDOW,
}
)
# Use Popen instead of run
process = subprocess.Popen(ffmpeg_cmd, **kwargs)
stdout, stderr = process.communicate() # Get stdout and stderr
# Check for errors in the ffmpeg command
if result.returncode != 0:
self.log_updated.emit((f"FFmpeg error: {result.stderr}", "red"))
if process.returncode != 0:
self.log_updated.emit((f"FFmpeg error (stderr):\n{stderr}", "red"))
if stdout: # Log stdout as well if it exists
self.log_updated.emit((f"FFmpeg output (stdout):\n{stdout}", "orange"))
# Log error but continue with original file
self.log_updated.emit(("Falling back to the original audio file without chapters\n", "orange"))
self.log_updated.emit(
("Falling back to the original audio file without chapters\n", "orange")
)
# Clean up chapters file even on error
if os.path.exists(chapters_info_path):
os.remove(chapters_info_path)
return out_path
else:
self.log_updated.emit(("Successfully added chapters to the audio file.\n", "green"))
self.log_updated.emit(
("Successfully added chapters to the audio file.\n", "green")
)
# delete the chapters path and original (wav) file
os.remove(chapters_info_path)
os.remove(out_path)
abogen/gui.py
+19 -8
View File
@@ -670,9 +670,16 @@ class abogen(QWidget):
format_layout = QVBoxLayout()
format_layout.addWidget(QLabel("Output Format:", self))
self.format_combo = QComboBox(self)
self.format_combo.setStyleSheet("QComboBox { min-height: 20px; padding: 6px 12px; }")
self.format_combo.setStyleSheet(
"QComboBox { min-height: 20px; padding: 6px 12px; }"
)
# Add items with display labels and underlying keys
for key, label in [("wav","wav"),("flac","flac"),("mp3","mp3"),("m4b","m4b (with chapters)")]:
for key, label in [
("wav", "wav"),
("flac", "flac"),
("mp3", "mp3"),
("m4b", "m4b (with chapters)"),
]:
self.format_combo.addItem(label, key)
# Initialize selection by matching saved key
idx = self.format_combo.findData(self.selected_format)
@@ -1422,6 +1429,7 @@ class abogen(QWidget):
if self.preview_playing:
try:
import pygame
pygame.mixer.music.stop()
except Exception:
pass
@@ -1436,18 +1444,20 @@ class abogen(QWidget):
self.voice_combo.setEnabled(False)
self.btn_voice_formula_mixer.setEnabled(False) # Disable mixer button
self.btn_start.setEnabled(False) # Disable start button during preview
# Start loading animation - ensure signal connection is always active
if hasattr(self, 'loading_movie'):
if hasattr(self, "loading_movie"):
# Disconnect previous connections to avoid multiple connections
try:
self.loading_movie.frameChanged.disconnect()
except TypeError:
pass # Ignore error if not connected
# Reconnect the signal
self.loading_movie.frameChanged.connect(
lambda: self.btn_preview.setIcon(QIcon(self.loading_movie.currentPixmap()))
lambda: self.btn_preview.setIcon(
QIcon(self.loading_movie.currentPixmap())
)
)
self.loading_movie.start()
@@ -1500,7 +1510,7 @@ class abogen(QWidget):
gpu_msg, gpu_ok = get_gpu_acceleration(self.use_gpu)
self.preview_thread = VoicePreviewThread(
np_module, kpipeline_class, lang, voice, speed, gpu_ok
np_module, kpipeline_class, lang, voice, speed, gpu_ok
)
self.preview_thread.finished.connect(self._play_preview_audio)
self.preview_thread.error.connect(self._preview_error)
@@ -1569,7 +1579,7 @@ class abogen(QWidget):
self.loading_movie.frameChanged.disconnect()
except Exception:
pass # Ignore error if not connected
self.btn_preview.setIcon(self.play_icon)
self.btn_preview.setIcon(self.play_icon)
self.btn_preview.setToolTip("Preview selected voice")
self.btn_preview.setEnabled(True)
self.voice_combo.setEnabled(True)
@@ -1845,6 +1855,7 @@ class abogen(QWidget):
def show_voice_formula_dialog(self):
from voice_profiles import load_profiles
profiles = load_profiles()
initial_state = None
selected_profile = self.selected_profile_name
abogen/utils.py
+1
View File
@@ -67,6 +67,7 @@ def get_version():
# Define config path
def get_user_config_path():
from platformdirs import user_config_dir
# TODO Config directory is changed for Linux and MacOS. But if old config exists, it will be used.
# On non‑Windows, prefer ~/.config/abogen if it already exists
if platform.system() != "Windows":
abogen/voice_formula_gui.py
+23 -11
View File
@@ -943,6 +943,7 @@ class VoiceFormulaDialog(QDialog):
def new_profile(self):
import re
while True:
name, ok = QInputDialog.getText(self, "New Profile", "Enter profile name:")
if not ok or not name:
@@ -950,8 +951,12 @@ class VoiceFormulaDialog(QDialog):
name = name.strip() # Remove leading/trailing spaces
if not name:
continue
if not re.match(r'^[\w\- ]+$', name):
QMessageBox.warning(self, "Invalid Name", "Profile name can only contain letters, numbers, spaces, underscores, and hyphens.")
if not re.match(r"^[\w\- ]+$", name):
QMessageBox.warning(
self,
"Invalid Name",
"Profile name can only contain letters, numbers, spaces, underscores, and hyphens.",
)
continue
profiles = load_profiles()
# Remove 'New profile' placeholder if not persisted in JSON
@@ -1173,13 +1178,16 @@ class VoiceFormulaDialog(QDialog):
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"):
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("*")
import re
while True:
new, ok = QInputDialog.getText(
self, "Rename Profile", f"Profile name:", text=old
@@ -1189,15 +1197,19 @@ class VoiceFormulaDialog(QDialog):
new = new.strip() # Remove leading/trailing spaces
if not new:
continue
if not re.match(r'^[\w\- ]+$', new):
QMessageBox.warning(self, "Invalid Name", "Profile name can only contain letters, numbers, spaces, underscores, and hyphens.")
if not re.match(r"^[\w\- ]+$", new):
QMessageBox.warning(
self,
"Invalid Name",
"Profile name can only contain letters, numbers, spaces, underscores, and hyphens.",
)
continue
profiles = load_profiles()
if new in profiles:
QMessageBox.warning(self, "Duplicate Name", "Profile already exists.")
continue
# Special case for renaming the virtual "New profile"
if self._virtual_new_profile and name == "New profile":
# Create the profile with the new name
@@ -1206,12 +1218,12 @@ class VoiceFormulaDialog(QDialog):
"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)
@@ -1220,11 +1232,11 @@ class VoiceFormulaDialog(QDialog):
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()
if hasattr(parent, "populate_profiles_in_voice_combo"):
parent.populate_profiles_in_voice_combo()