Overview and features Windows Speech Recognition

1 overview , features

1.1 interface

1.1.1 alternates panel


1.2 common commands

1.2.1 mousegrid
1.2.2 show numbers


1.3 dictation

1.3.1 speech dictionary


1.4 macros

overview , features

windows speech recognition allows user control computer, including operating system desktop user interface, through voice commands. applications, including of bundled windows, can controlled through voice commands. using speech recognition, users can dictate text within documents , e-mail messages, fill out forms, control operating system user interface, perform keyboard shortcuts, , move mouse cursor.

speech recognition uses speech profile store information user s voice. accuracy of speech recognition increases through use, helps feature adapt user s grammar, speech patterns, vocabulary, , word usage. speech recognition includes tutorial improve accuracy, , can optionally review user s personal documents, including e-mail messages, improve command , dictation accuracy. in windows 7 , later versions, additional option available allows users send speech information microsoft. individual speech profiles can created on per-user basis, , backups of profiles can performed via windows easy transfer or through downloable utility developed microsoft. profiles archived through utility carry wsrprofile filename extension. windows speech recognition relies on microsoft speech api. , third-party applications must support text services framework. speech recognition supports following languages: chinese (traditional), chinese (simplified), english (u.s.), english (u.k.), french, german, japanese, , spanish.

interface

the interface windows speech recognition consists of status area instructions, information commands (e.g., if command not heard speech recognizer), , information related state of speech recognizer; voice meter provided display visual feedback user voice volume levels. status area represents current state of windows speech recognition in total of 3 modes, listed below respective meanings:

listening: speech recognizer active , waiting user input
sleeping: speech recognizer not listen or respond commands other start listening
off: speech recognizer not listen or respond commands; mode can enabled speaking stop listening

in addition 3 modes listed above, status area can display information messages users can customize part of own windows speech recognition macros.

alternates panel

the alternates panel in windows speech recognition displaying suggestions phrase.

a disambiguation interface referred alternates panel displays list of items interpreted recognizer being relevant user s spoken word(s); if word or phrase user desired insert application listed among results, user can speak corresponding number of word appears among results , confirm choice speaking ok insert within application.

the alternates panel appear when launching programs or speaking commands may refer more 1 item (e.g., speaking start internet explorer may list web browser , alternate version of web browser add-ons disabled). however, windows registry entry, exactmatchoverpartialmatch, can limit commands programs or commands exact names if there more 1 instance of item included among results.

common commands

listed below common commands available windows speech recognition. words in italics indicate variable can substituted desired item (e.g., word direction in scroll direction command can substituted word down scroll down). start typing command enables windows speech recognition interpret dictation commands keyboard shortcuts.

dictation commands: new line, new paragraph, tab, literal word, numeral number, go word, go after word, no space, go start of sentence, go end of sentence, go start of paragraph, go end of paragraph, go start of document, go end of document, go field name (e.g., go address, cc, or subject). special characters, such comma, can dictated stating name of special character.
navigation commands:

keyboard shortcuts: press keyboard key, press ⇧ shift plus a, press capital b. nato phonetic alphabet supported. keys can pressed without first giving press command include: ← backspace, delete, end, ↵ enter, home, page down, page up, tab ↹.
mouse commands: click, click that, double-click, double-click that, mark, mark that, right-click, right-click that, mousegrid.
window management commands: close (alternatively maximize, minimize, or restore) window, close that, or close application name, switch applications, switch program name, scroll direction, scroll direction in number of pages, show desktop, show numbers.


speech recognition commands: start listening, stop listening, show speech options, open speech dictionary, move speech recognition, mimimize speech recognition. list of applicable commands can shown speaking can say? command available in english. users can query recognizer tasks in windows speaking how can task name, opens pane displays related information.

mousegrid

the mousegrid command overlaying numbers on windows vista desktop.

a mousegrid command enables users control mouse cursor overlaying numbers across 9 regions on screen; these regions narrow user speaks—by number(s)—which region focus on until reach desired interface element interact with. entire regions can interacted speaking click number of region, moves mouse cursor desired region , clicks it. individual item within region, such computer icon, can selected speaking mark number of region item appears. user can specify move marked item speaking click number of region. these commands work multiple regions of mousegrid.

show numbers

the show numbers command overlaying numbers on user interface elements in games explorer.

applications , operating system user interface elements not present obvious commands can still controlled asking system overlay numbers on top of them through show numbers command. once active, speaking overlaid number selects item user can open or perform other operations. command designed users interact items not readily identifiable.

dictation

windows speech recognition enables dictation of text in operating system , applications. applications not automatically support dictation, option enable dictation everywhere available. if mistake in dictation occurs, user can correct mistake saying correct word or correct , alternate panel interface appear , provide suggestions correction; these suggestions can selected speaking number corresponding number of suggestion in list , speaking ok. if desired word not listed among included suggestions, user can speak desired word might appear. alternatively, spell command allows user speak desired word on per-letter basis appear among suggestions. multiple words in sentence may corrected @ time. example, if user states dictating speech recognition recognizes word thing, user can state correct thing correct both words.

speech dictionary

wsr includes personal dictionary allows users include or exclude words or expressions being dictated. default, dictionary includes on 100,000 words in english language. when user adds word beginning capital letter dictionary, user can specify whether should capitalized during dictation or if capitalization depends on context word spoken; users may record pronunciations words added dictionary increase recognition accuracy; words written via stylus on tablet pc windows handwriting recognition feature stored. of information stored within dictionary included part of user s speech profile.

macros

an aero wizard interface displaying options create speech recognition macros.

windows speech recognition supports custom macros through separate utility released microsoft enables use of commands further based on natural language processing. example of functionality, e-mail macro released microsoft enables natural language command user can state send e-mail contact subject, opens microsoft outlook compose new message designated contact , subject automatically inserted within application. microsoft has released sample macros speech dictionary, windows media player, microsoft powerpoint, speech synthesis, switch between multiple microphones, customize various aspects of audio device configuration such volume levels, , general natural language queries such as, weather forecast? time it? , s date? answers these queries spoken user via speech synthesizer.

users , developers can create own custom macros can based on text transcription , substitution, program execution (with support command-line arguments), keyboard shortcuts, emulation of existing voice commands, or combination of these items. xml, jscript , vbscript supported. macros can limited individual applications if desired, , rules macros can defined programmatically.

in order macro loaded, must stored within speech macros folder within current user s documents directory. default, macros digitally signed if user certificate available in order ensure created commands not loaded or tampered third-parties; if 1 not available, administrator can create certificate use. macros utility includes security levels prohibit unsigned macros being loaded, prompt users sign macros, , load unsigned macros if user desires occur.