feat: Enhance normalization settings with additional options and UI elements

abogen/web/routes.py

@@ -1778,10 +1778,17 @@ def _apply_book_step_form(
         return default
 
     overrides_existing = getattr(pending, "normalization_overrides", None)
-    overrides: Dict[str, bool] = dict(overrides_existing or {})
+    overrides: Dict[str, Any] = dict(overrides_existing or {})
-    for key in _APOSTROPHE_OVERRIDE_KEYS:
+    for key in _NORMALIZATION_BOOLEAN_KEYS:
         default_toggle = overrides.get(key, bool(settings.get(key, True)))
         overrides[key] = _extract_checkbox(key, default_toggle)
+    for key in _NORMALIZATION_STRING_KEYS:
+        default_val = overrides.get(key, str(settings.get(key, "")))
+        val = form.get(key)
+        if val is not None:
+            overrides[key] = str(val)
+        else:
+            overrides[key] = default_val
     pending.normalization_overrides = overrides
 
     speed_value = form.get("speed")
@@ -2014,6 +2021,7 @@ def _template_options() -> Dict[str, Any]:
         "speaker_pronunciation_sentence": current_settings.get(
             "speaker_pronunciation_sentence", _settings_defaults()["speaker_pronunciation_sentence"]
         ),
+        "apostrophe_modes": _APOSTROPHE_MODE_OPTIONS,
     }
 
 
@@ -2112,7 +2120,12 @@ BOOLEAN_SETTINGS = {
 FLOAT_SETTINGS = {"silence_between_chapters", "chapter_intro_delay", "llm_timeout"}
 INT_SETTINGS = {"max_subtitle_words", "speaker_analysis_threshold"}
 
-_APOSTROPHE_OVERRIDE_KEYS = (
+_NORMALIZATION_BOOLEAN_KEYS = (
+    "normalization_numbers",
+    "normalization_titles",
+    "normalization_terminal",
+    "normalization_phoneme_hints",
+    "normalization_caps_quotes",
     "normalization_apostrophes_contractions",
     "normalization_apostrophes_plural_possessives",
     "normalization_apostrophes_sibilant_possessives",
@@ -2124,7 +2137,11 @@ _APOSTROPHE_OVERRIDE_KEYS = (
     "normalization_contraction_modal_would",
     "normalization_contraction_negation_not",
     "normalization_contraction_let_us",
-    "normalization_caps_quotes",
+)
+
+_NORMALIZATION_STRING_KEYS = (
+    "normalization_numbers_year_style",
+    "normalization_apostrophe_mode",
 )
 
 
@@ -4431,7 +4448,8 @@ def _build_pending_job_from_extraction(
         metadata_tags=metadata_tags,
         chapters=chapters_payload,
         normalization_overrides={
-            key: bool(settings.get(key, True)) for key in _APOSTROPHE_OVERRIDE_KEYS
+            **{key: bool(settings.get(key, True)) for key in _NORMALIZATION_BOOLEAN_KEYS},
+            **{key: str(settings.get(key, "")) for key in _NORMALIZATION_STRING_KEYS},
         },
         created_at=time.time(),
         cover_image_path=cover_path,

abogen/web/service.py

@@ -1095,10 +1095,13 @@ class ConversionService:
 
         if existing_items:
             job.add_log(
-                f"Audiobookshelf upload skipped: '{display_title}' already exists in the configured folder.",
+                f"Removing existing Audiobookshelf item(s) for '{display_title}' before upload.",
-                level="warning",
+                level="info",
             )
-            return
+            try:
+                client.delete_items(existing_items)
+            except Exception as exc:
+                job.add_log(f"Failed to remove existing item(s): {exc}", level="warning")
 
         client.upload_audiobook(
             audio_path,

abogen/web/templates/partials/new_job_step_book.html

@@ -128,6 +128,26 @@
 {% endif %}
 {% set normalization_overrides = pending.normalization_overrides if pending and pending.normalization_overrides else {} %}
 {% set normalization_toggles = [
+  {
+    'name': 'normalization_numbers',
+    'label': "Convert grouped numbers to words",
+    'value': normalization_overrides.get('normalization_numbers', settings_dict.get('normalization_numbers', True)),
+  },
+  {
+    'name': 'normalization_titles',
+    'label': "Expand titles and suffixes (Dr., St., Jr., …)",
+    'value': normalization_overrides.get('normalization_titles', settings_dict.get('normalization_titles', True)),
+  },
+  {
+    'name': 'normalization_terminal',
+    'label': "Ensure sentences end with terminal punctuation",
+    'value': normalization_overrides.get('normalization_terminal', settings_dict.get('normalization_terminal', True)),
+  },
+  {
+    'name': 'normalization_phoneme_hints',
+    'label': "Apply phoneme hints for common abbreviations",
+    'value': normalization_overrides.get('normalization_phoneme_hints', settings_dict.get('normalization_phoneme_hints', True)),
+  },
   {
     'name': 'normalization_caps_quotes',
     'label': "Convert ALL CAPS dialogue inside quotes",
@@ -158,6 +178,36 @@
     'label': "Expand leading elisions ('tis -> it is)",
     'value': normalization_overrides.get('normalization_apostrophes_leading_elisions', settings_dict.get('normalization_apostrophes_leading_elisions', True)),
   },
+  {
+    'name': 'normalization_contraction_aux_be',
+    'label': "Expand auxiliary 'be' (I'm -> I am)",
+    'value': normalization_overrides.get('normalization_contraction_aux_be', settings_dict.get('normalization_contraction_aux_be', True)),
+  },
+  {
+    'name': 'normalization_contraction_aux_have',
+    'label': "Expand auxiliary 'have' (I've -> I have)",
+    'value': normalization_overrides.get('normalization_contraction_aux_have', settings_dict.get('normalization_contraction_aux_have', True)),
+  },
+  {
+    'name': 'normalization_contraction_modal_will',
+    'label': "Expand modal 'will' (I'll -> I will)",
+    'value': normalization_overrides.get('normalization_contraction_modal_will', settings_dict.get('normalization_contraction_modal_will', True)),
+  },
+  {
+    'name': 'normalization_contraction_modal_would',
+    'label': "Expand modal 'would' (I'd -> I would)",
+    'value': normalization_overrides.get('normalization_contraction_modal_would', settings_dict.get('normalization_contraction_modal_would', True)),
+  },
+  {
+    'name': 'normalization_contraction_negation_not',
+    'label': "Expand negation 'not' (don't -> do not)",
+    'value': normalization_overrides.get('normalization_contraction_negation_not', settings_dict.get('normalization_contraction_negation_not', True)),
+  },
+  {
+    'name': 'normalization_contraction_let_us',
+    'label': "Expand 'let's' -> let us",
+    'value': normalization_overrides.get('normalization_contraction_let_us', settings_dict.get('normalization_contraction_let_us', True)),
+  },
 ] %}
 {% set voice_formula_value = '' %}
 {% set profile_value = narrator_profile if narrator_profile else '__standard' %}
@@ -303,6 +353,24 @@
     <section class="form-section">
       <h3 class="form-section__title">Text normalization</h3>
       <div class="field-grid field-grid--compact">
+        <div class="field field--stack field--span-2">
+          <span class="field__label">Apostrophe strategy</span>
+          <div class="choices choices--inline">
+            {% for option in options.apostrophe_modes %}
+            <label class="radio-pill">
+              <input type="radio" name="normalization_apostrophe_mode" value="{{ option.value }}" {% if normalization_overrides.get('normalization_apostrophe_mode', settings_dict.get('normalization_apostrophe_mode', 'spacy')) == option.value %}checked{% endif %} {{ 'disabled' if readonly else '' }}>
+              <span>{{ option.label }}</span>
+            </label>
+            {% endfor %}
+          </div>
+        </div>
+        <div class="field field--stack field--span-2">
+          <label for="normalization_numbers_year_style">Year pronunciation style</label>
+          <select id="normalization_numbers_year_style" name="normalization_numbers_year_style" {{ 'disabled' if readonly else '' }}>
+            <option value="american" {% if normalization_overrides.get('normalization_numbers_year_style', settings_dict.get('normalization_numbers_year_style', 'american')) == 'american' %}selected{% endif %}>American (1990 -> nineteen ninety)</option>
+            <option value="off" {% if normalization_overrides.get('normalization_numbers_year_style', settings_dict.get('normalization_numbers_year_style', 'american')) == 'off' %}selected{% endif %}>Off (read as number)</option>
+          </select>
+        </div>
   {% for toggle in normalization_toggles %}
         <div class="field field--stack">
           <label class="toggle-pill">