Add dictation command processing for voice input

build.gradle

@@ -328,6 +328,10 @@ android {
             res.srcDirs = ['java/unstable/res', translationsWithoutEngValues('translations/devbuild')]
         }
 
+        test {
+            java.srcDirs = ['java/test']
+        }
+
         androidTest {
             res.srcDirs = ['tests/res']
             java.srcDirs = ['tests/src']

java/res/values/strings-uix.xml

@@ -538,6 +538,26 @@
     <string name="voice_input_settings_change_models">Models</string>
     <string name="voice_input_settings_change_models_subtitle">To change the models, visit Languages &amp; Models menu</string>
 
+    <!-- Dictation commands settings -->
+    <string name="dictation_commands_title">Dictation Commands</string>
+    <string name="dictation_commands_subtitle">Replace spoken phrases like \"new line\" or \"dollar sign\" with symbols and formatting</string>
+    <string name="dictation_formatting_title">Formatting</string>
+    <string name="dictation_formatting_subtitle">New line, new paragraph, tab, numeral, no space on/off</string>
+    <string name="dictation_capitalization_title">Capitalization</string>
+    <string name="dictation_capitalization_subtitle">Caps on/off, all caps</string>
+    <string name="dictation_punctuation_title">Punctuation &amp; Brackets</string>
+    <string name="dictation_punctuation_subtitle">Quotes, brackets, parentheses, dash, ellipsis</string>
+    <string name="dictation_symbols_title">Symbols</string>
+    <string name="dictation_symbols_subtitle">Ampersand, asterisk, at sign, hashtag, etc.</string>
+    <string name="dictation_math_title">Math Symbols</string>
+    <string name="dictation_math_subtitle">Equal, plus, minus, greater than, less than</string>
+    <string name="dictation_currency_title">Currency Symbols</string>
+    <string name="dictation_currency_subtitle">Dollar, euro, pound, yen, cent</string>
+    <string name="dictation_emoticons_title">Emoticons</string>
+    <string name="dictation_emoticons_subtitle">Smiley, frowny, winky face</string>
+    <string name="dictation_ip_marks_title">Intellectual Property Marks</string>
+    <string name="dictation_ip_marks_subtitle">Copyright, registered, trademark</string>
+
     <!-- Prediction menu -->
     <string name="prediction_settings_title">Text Prediction</string>
     <string name="prediction_settings_transformer">Transformer LM</string>

java/src/org/futo/inputmethod/latin/uix/VoiceInputSettingKeys.kt

@@ -67,4 +67,50 @@ val LANGUAGE_TOGGLES = SettingsKey(
 val USE_PERSONAL_DICT = SettingsKey(
     key = booleanPreferencesKey("use_personal_dict_voice_input"),
     default = true
+)
+
+// Dictation command settings
+val DICTATION_COMMANDS_ENABLED = SettingsKey(
+    key = booleanPreferencesKey("dictation_commands_enabled"),
+    default = true
+)
+
+val DICTATION_FORMATTING = SettingsKey(
+    key = booleanPreferencesKey("dictation_formatting"),
+    default = true
+)
+
+val DICTATION_CAPITALIZATION = SettingsKey(
+    key = booleanPreferencesKey("dictation_capitalization"),
+    default = true
+)
+
+val DICTATION_PUNCTUATION = SettingsKey(
+    key = booleanPreferencesKey("dictation_punctuation"),
+    default = true
+)
+
+val DICTATION_SYMBOLS = SettingsKey(
+    key = booleanPreferencesKey("dictation_symbols"),
+    default = true
+)
+
+val DICTATION_MATH = SettingsKey(
+    key = booleanPreferencesKey("dictation_math"),
+    default = true
+)
+
+val DICTATION_CURRENCY = SettingsKey(
+    key = booleanPreferencesKey("dictation_currency"),
+    default = true
+)
+
+val DICTATION_EMOTICONS = SettingsKey(
+    key = booleanPreferencesKey("dictation_emoticons"),
+    default = true
+)
+
+val DICTATION_IP_MARKS = SettingsKey(
+    key = booleanPreferencesKey("dictation_ip_marks"),
+    default = true
 )

java/src/org/futo/inputmethod/latin/uix/actions/VoiceInputAction.kt

@@ -25,11 +25,20 @@ import kotlinx.coroutines.launch
 import kotlinx.coroutines.runBlocking
 import kotlinx.coroutines.yield
 import org.futo.inputmethod.latin.R
-import org.futo.inputmethod.latin.uix.AUDIO_FOCUS
 import org.futo.inputmethod.latin.uix.Action
 import org.futo.inputmethod.latin.uix.ActionWindow
+import org.futo.inputmethod.latin.uix.AUDIO_FOCUS
 import org.futo.inputmethod.latin.uix.CAN_EXPAND_SPACE
 import org.futo.inputmethod.latin.uix.CloseResult
+import org.futo.inputmethod.latin.uix.DICTATION_CAPITALIZATION
+import org.futo.inputmethod.latin.uix.DICTATION_COMMANDS_ENABLED
+import org.futo.inputmethod.latin.uix.DICTATION_CURRENCY
+import org.futo.inputmethod.latin.uix.DICTATION_EMOTICONS
+import org.futo.inputmethod.latin.uix.DICTATION_FORMATTING
+import org.futo.inputmethod.latin.uix.DICTATION_IP_MARKS
+import org.futo.inputmethod.latin.uix.DICTATION_MATH
+import org.futo.inputmethod.latin.uix.DICTATION_PUNCTUATION
+import org.futo.inputmethod.latin.uix.DICTATION_SYMBOLS
 import org.futo.inputmethod.latin.uix.DISALLOW_SYMBOLS
 import org.futo.inputmethod.latin.uix.ENABLE_SOUND
 import org.futo.inputmethod.latin.uix.KeyboardManagerForAction
@@ -42,6 +51,8 @@ import org.futo.inputmethod.latin.uix.VERBOSE_PROGRESS
 import org.futo.inputmethod.latin.uix.getSetting
 import org.futo.inputmethod.latin.uix.setSetting
 import org.futo.inputmethod.latin.uix.settings.SettingsActivity
+import org.futo.inputmethod.latin.uix.utils.DictationCommandProcessor
+import org.futo.inputmethod.latin.uix.utils.DictationSettings
 import org.futo.inputmethod.latin.uix.utils.ModelOutputSanitizer
 import org.futo.inputmethod.latin.xlm.UserDictionaryObserver
 import org.futo.inputmethod.updates.openURI
@@ -118,6 +129,7 @@ private class VoiceInputActionWindow(
     val context = manager.getContext()
 
     private var shouldPlaySounds: Boolean = false
+    private var dictationSettings: DictationSettings = DictationSettings()
     private fun loadSettings(): RecognizerViewSettings {
         val enableSound = context.getSetting(ENABLE_SOUND)
         val verboseFeedback = false//context.getSetting(VERBOSE_PROGRESS)
@@ -139,6 +151,18 @@ private class VoiceInputActionWindow(
 
         shouldPlaySounds = enableSound
 
+        dictationSettings = DictationSettings(
+            enabled = context.getSetting(DICTATION_COMMANDS_ENABLED),
+            formatting = context.getSetting(DICTATION_FORMATTING),
+            capitalization = context.getSetting(DICTATION_CAPITALIZATION),
+            punctuation = context.getSetting(DICTATION_PUNCTUATION),
+            symbols = context.getSetting(DICTATION_SYMBOLS),
+            math = context.getSetting(DICTATION_MATH),
+            currency = context.getSetting(DICTATION_CURRENCY),
+            emoticons = context.getSetting(DICTATION_EMOTICONS),
+            ipMarks = context.getSetting(DICTATION_IP_MARKS)
+        )
+
         return RecognizerViewSettings(
             shouldShowInlinePartialResult = false,
             shouldShowVerboseFeedback = verboseFeedback,
@@ -263,8 +287,10 @@ private class VoiceInputActionWindow(
         wasFinished = true
 
         manager.getLifecycleScope().launch(Dispatchers.Main) {
+            // Sanitize first so dictation formatting chars aren't mangled by trim()
             val sanitized = ModelOutputSanitizer.sanitize(result, inputTransaction.textContext)
-            inputTransaction.commit(sanitized)
+            val processed = DictationCommandProcessor.process(sanitized, dictationSettings)
+            inputTransaction.commit(processed)
             manager.announce(result)
             manager.closeActionWindow()
         }
@@ -273,7 +299,8 @@ private class VoiceInputActionWindow(
     override fun partialResult(result: String) {
         manager.getLifecycleScope().launch(Dispatchers.Main) {
             val sanitized = ModelOutputSanitizer.sanitize(result, inputTransaction.textContext)
-            inputTransaction.updatePartial(sanitized)
+            val processed = DictationCommandProcessor.process(sanitized, dictationSettings)
+            inputTransaction.updatePartial(processed)
         }
     }
 

java/src/org/futo/inputmethod/latin/uix/settings/SettingsNavigator.kt

@@ -55,6 +55,7 @@ import org.futo.inputmethod.latin.uix.settings.pages.SelectLanguageScreen
 import org.futo.inputmethod.latin.uix.settings.pages.SelectLayoutsScreen
 import org.futo.inputmethod.latin.uix.settings.pages.themes.ThemeScreen
 import org.futo.inputmethod.latin.uix.settings.pages.TypingSettingsMenu
+import org.futo.inputmethod.latin.uix.settings.pages.DictationCommandsMenu
 import org.futo.inputmethod.latin.uix.settings.pages.VoiceInputMenu
 import org.futo.inputmethod.latin.uix.settings.pages.addModelManagerNavigation
 import org.futo.inputmethod.latin.uix.settings.pages.buggyeditors.BuggyTextEditVariations
@@ -86,6 +87,7 @@ val SettingsMenus = listOf(
     PredictiveTextMenu,
     BlacklistScreenLite,
     VoiceInputMenu,
+    DictationCommandsMenu,
     ActionsScreen,
     HelpMenu,
     MiscMenu,

java/src/org/futo/inputmethod/latin/uix/settings/pages/VoiceInput.kt

@@ -5,6 +5,15 @@ import androidx.compose.runtime.Composable
 import org.futo.inputmethod.latin.R
 import org.futo.inputmethod.latin.uix.AUDIO_FOCUS
 import org.futo.inputmethod.latin.uix.CAN_EXPAND_SPACE
+import org.futo.inputmethod.latin.uix.DICTATION_CAPITALIZATION
+import org.futo.inputmethod.latin.uix.DICTATION_COMMANDS_ENABLED
+import org.futo.inputmethod.latin.uix.DICTATION_CURRENCY
+import org.futo.inputmethod.latin.uix.DICTATION_EMOTICONS
+import org.futo.inputmethod.latin.uix.DICTATION_FORMATTING
+import org.futo.inputmethod.latin.uix.DICTATION_IP_MARKS
+import org.futo.inputmethod.latin.uix.DICTATION_MATH
+import org.futo.inputmethod.latin.uix.DICTATION_PUNCTUATION
+import org.futo.inputmethod.latin.uix.DICTATION_SYMBOLS
 import org.futo.inputmethod.latin.uix.DISALLOW_SYMBOLS
 import org.futo.inputmethod.latin.uix.ENABLE_SOUND
 import org.futo.inputmethod.latin.uix.PREFER_BLUETOOTH
@@ -22,6 +31,66 @@ private val visibilityCheckNotSystemVoiceInput = @Composable {
     useDataStoreValue(USE_SYSTEM_VOICE_INPUT) == false
 }
 
+val DictationCommandsMenu = UserSettingsMenu(
+    title = R.string.dictation_commands_title,
+    navPath = "dictationCommands", registerNavPath = true,
+    settings = listOf(
+        userSettingToggleDataStore(
+            title = R.string.dictation_commands_title,
+            subtitle = R.string.dictation_commands_subtitle,
+            setting = DICTATION_COMMANDS_ENABLED
+        ),
+
+        userSettingToggleDataStore(
+            title = R.string.dictation_formatting_title,
+            subtitle = R.string.dictation_formatting_subtitle,
+            setting = DICTATION_FORMATTING
+        ),
+
+        userSettingToggleDataStore(
+            title = R.string.dictation_capitalization_title,
+            subtitle = R.string.dictation_capitalization_subtitle,
+            setting = DICTATION_CAPITALIZATION
+        ),
+
+        userSettingToggleDataStore(
+            title = R.string.dictation_punctuation_title,
+            subtitle = R.string.dictation_punctuation_subtitle,
+            setting = DICTATION_PUNCTUATION
+        ),
+
+        userSettingToggleDataStore(
+            title = R.string.dictation_symbols_title,
+            subtitle = R.string.dictation_symbols_subtitle,
+            setting = DICTATION_SYMBOLS
+        ),
+
+        userSettingToggleDataStore(
+            title = R.string.dictation_math_title,
+            subtitle = R.string.dictation_math_subtitle,
+            setting = DICTATION_MATH
+        ),
+
+        userSettingToggleDataStore(
+            title = R.string.dictation_currency_title,
+            subtitle = R.string.dictation_currency_subtitle,
+            setting = DICTATION_CURRENCY
+        ),
+
+        userSettingToggleDataStore(
+            title = R.string.dictation_emoticons_title,
+            subtitle = R.string.dictation_emoticons_subtitle,
+            setting = DICTATION_EMOTICONS
+        ),
+
+        userSettingToggleDataStore(
+            title = R.string.dictation_ip_marks_title,
+            subtitle = R.string.dictation_ip_marks_subtitle,
+            setting = DICTATION_IP_MARKS
+        )
+    )
+)
+
 val VoiceInputMenu = UserSettingsMenu(
     title = R.string.voice_input_settings_title,
     navPath = "voiceInput", registerNavPath = true,
@@ -82,6 +151,13 @@ val VoiceInputMenu = UserSettingsMenu(
             setting = USE_VAD_AUTOSTOP
         ).copy(visibilityCheck = visibilityCheckNotSystemVoiceInput),
 
+        userSettingNavigationItem(
+            title = R.string.dictation_commands_title,
+            subtitle = R.string.dictation_commands_subtitle,
+            style = NavigationItemStyle.Misc,
+            navigateTo = "dictationCommands"
+        ).copy(visibilityCheck = visibilityCheckNotSystemVoiceInput),
+
         userSettingNavigationItem(
             title = R.string.voice_input_settings_change_models,
             subtitle = R.string.voice_input_settings_change_models_subtitle,