v1.0.6

CHANGELOG.md

@@ -1,3 +1,5 @@
-- 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.
+- Added `Insert chapter marker` button in text editor to insert chapter markers at the current cursor position.
+- Added `Preview` button in voice mixer to preview the voice mix with the selected settings.
+- Fixed `f-string: unmatched '['` error in Voice preview, mentioned in #14
+- Fixed the issue with the content before first chapter not being included in the output.
+- Fixed m4b chapter generation opens CMD window in Windows.

abogen/VERSION

@@ -1 +1 @@
-1.0.5
+1.0.6

abogen/conversion.py

@@ -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"])
@@ -224,6 +228,12 @@ class ConversionThread(QThread):
             chapter_splits = list(re.finditer(chapter_pattern, text))
             chapters = []
             if chapter_splits:
+                # prepend Introduction for content before first marker
+                first_start = chapter_splits[0].start()
+                if first_start > 0:
+                    intro_text = text[:first_start].strip()
+                    if intro_text:
+                        chapters.append(("Introduction", intro_text))
                 for idx, match in enumerate(chapter_splits):
                     start = match.end()
                     end = (
@@ -610,45 +620,86 @@ 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")
                 # use 10th of second for start/end time
-                f.write(f"START={int(chapter["start"]*10)}\n")
-                f.write(f"END={int(chapter["end"]*10)}\n")
-                f.write(f"title={chapter["chapter"]}\n\n")
+                f.write(f"START={int(chapter['start']*10)}\n")
+                f.write(f"END={int(chapter['end']*10)}\n")
+                f.write(f"title={chapter['chapter']}\n\n")
         # 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)

abogen/gui.py

@@ -290,8 +290,8 @@ class InputBox(QLabel):
                 )
                 win.selected_book_path = file_path
                 win.open_book_file(
-                    file_path
-                )  # This will handle the dialog and setting file info
+                    file_path  # This will handle the dialog and setting file info
+                )
                 event.acceptProposedAction()
             else:
                 self.set_error("Please drop a .txt, .epub, or .pdf file.")
@@ -328,7 +328,7 @@ class TextboxDialog(QDialog):
         self.setWindowFlags(
             Qt.Window | Qt.WindowCloseButtonHint | Qt.WindowMaximizeButtonHint
         )
-        self.resize(600, 400)
+        self.resize(700, 500)
 
         layout = QVBoxLayout(self)
 
@@ -358,6 +358,11 @@ class TextboxDialog(QDialog):
         self.save_as_button.clicked.connect(self.save_as_text)
         self.save_as_button.setToolTip("Save the current text to a file")
 
+        self.insert_chapter_btn = QPushButton("Insert Chapter Marker", self)
+        self.insert_chapter_btn.setToolTip("Insert a chapter marker at the cursor")
+        self.insert_chapter_btn.clicked.connect(self.insert_chapter_marker)
+        button_layout.addWidget(self.insert_chapter_btn)
+
         self.cancel_button = QPushButton("Cancel", self)
         self.cancel_button.clicked.connect(self.reject)
 
@@ -444,6 +449,13 @@ class TextboxDialog(QDialog):
         except Exception as e:
             QMessageBox.critical(self, "Save Error", f"Could not save file:\n{e}")
 
+    def insert_chapter_marker(self):
+        # Insert a fixed chapter marker without prompting
+        cursor = self.text_edit.textCursor()
+        cursor.insertText("<<CHAPTER_MARKER:Title>>")
+        self.text_edit.setTextCursor(cursor)
+        self.update_char_count()
+
 
 class abogen(QWidget):
     def __init__(self):
@@ -670,9 +682,16 @@ 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 +1441,7 @@ class abogen(QWidget):
         if self.preview_playing:
             try:
                 import pygame
+
                 pygame.mixer.music.stop()
             except Exception:
                 pass
@@ -1438,7 +1458,7 @@ class abogen(QWidget):
         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()
@@ -1447,7 +1467,9 @@ class abogen(QWidget):
 
             # 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()
 
@@ -1478,14 +1500,14 @@ class abogen(QWidget):
             # Build voice formula string
             components = [f"{name}*{weight}" for name, weight in self.mixed_voice_state]
             voice = " + ".join(filter(None, components))
-            # determine language: use profile setting if available, else first voice code
+            # determine language: use profile setting, else explicit mixer selection, else fallback to first voice code
             if self.selected_profile_name:
                 from voice_profiles import load_profiles
 
                 entry = load_profiles().get(self.selected_profile_name, {})
                 lang = entry.get("language")
             else:
-                lang = None
+                lang = self.selected_lang
             if not lang and self.mixed_voice_state:
                 lang = (
                     self.mixed_voice_state[0][0][0]
@@ -1845,6 +1867,7 @@ 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

@@ -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

@@ -47,7 +47,7 @@ VOICE_MIXER_WIDTH = 100
 SLIDER_WIDTH = 32
 MIN_WINDOW_WIDTH = 600
 MIN_WINDOW_HEIGHT = 400
-INITIAL_WINDOW_WIDTH = 1000
+INITIAL_WINDOW_WIDTH = 1200
 INITIAL_WINDOW_HEIGHT = 500
 
 # Language options for the language selector loaded from constants
@@ -413,6 +413,11 @@ class VoiceFormulaDialog(QDialog):
             self.language_combo.setCurrentIndex(idx)
         self.language_combo.currentIndexChanged.connect(self.mark_profile_modified)
         header_row.addWidget(self.language_combo)
+        # Preview current voice mix using main window's preview
+        self.btn_preview_mix = QPushButton("Preview", self)
+        self.btn_preview_mix.setToolTip("Preview current voice mix")
+        self.btn_preview_mix.clicked.connect(self.preview_current_mix)
+        header_row.addWidget(self.btn_preview_mix)
         mixer_layout.addLayout(header_row)
 
         # Error message
@@ -708,6 +713,9 @@ class VoiceFormulaDialog(QDialog):
         ]
 
         total = sum(w for _, w in selected)
+        # disable Preview if no voices selected, but don't enable while loading
+        if not getattr(self, "_loading", False):
+            self.btn_preview_mix.setEnabled(total > 0)
 
         if total > 0:
             self.error_label.hide()
@@ -943,6 +951,7 @@ 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 +959,12 @@ 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 +1186,16 @@ 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
@@ -1189,8 +1205,12 @@ 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()
@@ -1326,3 +1346,39 @@ class VoiceFormulaDialog(QDialog):
                 else:
                     item.setBackground(QColor("white"))
         self.update_profile_save_buttons()
+
+    def preview_current_mix(self):
+        # Disable preview until playback completes
+        self.btn_preview_mix.setEnabled(False)
+        self.btn_preview_mix.setText("Loading...")
+        self._loading = True
+        parent = self.parent()
+        if parent and hasattr(parent, "preview_voice"):
+            # Apply mixed voices and selected language
+            parent.mixed_voice_state = self.get_selected_voices()
+            parent.selected_profile_name = None
+            lang = self.language_combo.currentData()
+            parent.selected_lang = lang
+            parent.subtitle_combo.setEnabled(
+                lang in SUPPORTED_LANGUAGES_FOR_SUBTITLE_GENERATION
+            )
+            # Reset start flag and trigger preview
+            self._started = False
+            parent.preview_voice()
+            # Poll preview_playing: wait for start then end
+            self._preview_poll_timer = QTimer(self)
+            self._preview_poll_timer.timeout.connect(self._check_preview_done)
+            self._preview_poll_timer.start(200)
+
+    def _check_preview_done(self):
+        parent = self.parent()
+        if parent and hasattr(parent, "preview_playing"):
+            # Mark when playback starts
+            if parent.preview_playing:
+                self._started = True
+            # Once started and then stopped, re-enable
+            elif getattr(self, "_started", False):
+                self.btn_preview_mix.setEnabled(True)
+                self.btn_preview_mix.setText("Preview")
+                self._loading = False
+                self._preview_poll_timer.stop()