Custom Voice Commands
Last updated
Last updated
Custom voice commands let you map a phrase to another phrase or command. This powerful feature can be used to correct common grammatical errors, avoid repetitive data entry, trigger commands, insert uncommon words, and more.
In this document, we will go over how to create custom voice commands, create voice commands that match within words, action commands, and conclude with some popular use cases for custom voice commands.
a. b. c. d.
If you want to use voice commands to perform an action like changing the dictation language, pressing a keyboard shortcut, opening a webpage, closing the current tab, or simply inserting a line break, go to the section.
Step 1: Ensure you have the extension installed and have upgraded to .
Step 2: Open the .
Step 3: Go to the Custom Commands section and click Add Command.
A pop-up on the page asks for the custom voice command details.
Say This - phrase that will activate this voice command &#xNAN;To Insert This - phrase you want to insert
For example, say you want to add a voice command for your phone number. You can do something like below:
Fill out the form and click the Add new command button. In the example above, saying "insert phone" will insert "217-652-8632".
You can add multiple commands at once using Bulk Add.
This will open up a pop-up. You can enter a list of voice commands in CSV format and bulk-add them.
By default, voice commands do whole-word matching. If you have a command for the word "magic", it will only match "magic" but not "magical". To match both "magic" and "magical" using the phrase "magic", you need to check the match within the word option when creating the voice command.
Action commands are commands that accomplish a specific task. They generally do not insert any text. Voice In supports the following action commands:
Command
Action
<newline>
Insert a line break
<newparagraph>
Insert a paragraph break
<paste>
Paste text from the clipboard
<undo>
Undo the last text insertion
<deleteword>
Delete last word
<deleteword:n>
Delete last n words. (See section e below)
<capson>
Insert text in upper case going forward (behaves as caps lock on)
<capsoff>
Turns off <capson>
<capitalize>
Capitalize the next word
<enter>
Press enter key
<tab>
Press tab key
<delete>
Delete selection
<open:*>
Open a webpage. Example <open:https://google.com> (See section a below)
<press:*>
Press a keyboard shortcut. Example <press:shift+enter> (See section b below)
<scrollup>,<scrolldown>, <scrolltop>,<scrollbottom>
Scroll on the current page up, down, to the top, and the bottom respectively
<closetab>
Close the current tab
<lang:*>
Change language. Example <lang:es-ES> to change to Spanish (Spain) (See section c below)
<case:*>
Change dictation case. Example <case:upper> to use uppercase (See section d below)
<click:*>
Click a link or button with given text. (See section f below)
<date>
Insert current date
<date:*>
Insert current date and time in a specified format. Example: <date:MM/dd/yyyy hh:mm a>. (See section g below)
<unset>
Use this to remove a built-in voice command. Example: period => <unset> will remove the period built-in voice command.
You can add a voice command mapping any phrase to the action commands listed above.
For example, a common dictation error when you say "new line" is "uline". To trigger a new line even when it is incorrectly dictated as uline, you can add the voice command: uline => <newline>
Some of the action commands are configurable. For example, you can add a voice command use spanish => <lang:es-ES> to quickly change the dictation language to Spanish.
There are 5 configurable action commands <open:*>, <press:*>, <lang*>, <case:*> and <deleteword:n>
Use this action command to open a webpage. Example: open google => <open:https://google.com>
Use this action command to trigger a key press. For example, in Gmail, the shortcut to send the email is ctrl + enter (cmd + enter on Mac). You can trigger this action using voice commands by adding a voice command "send mail" => <press:ctrl+enter> (or <press:cmd+enter> on Mac).
Voice In supports transforming dictated text to a specific case. The five supported cases are default, lowercase, UPPERCASE, Sentence case, and Capitalize Each Word. Use this <case:*> action command to switch between them. Example: uppercase => <case:upper> Possible values are <case:default>, <case:lower>, <case:upper>, <case:sentence>, <case:capitalize> for default, lowercase, UPPERCASE, Sentence case, and Capitalize Each Word, respectively.
Use this action command to delete the last n words. Example: delete last two words => <deleteword:2>
Use this action to click a link or button on the page. Example: click submit ⇒ <click:submit>
If there are multiple links/buttons with the given text, the first one with the text will be clicked.
Use this voice command to insert the current date and time in a specified format. Example: insert date ⇒ <date:MM/dd/yyyy hh:mm a>
The above command would insert the date as 05/15/2025 02:30 pm
The formatting options supported are:
To remove a built-in voice command in Voice In, use the <unset> action in a custom command. For instance, if you want to remove the built-in voice command "period" that inserts a dot (.), you can create a custom voice command like this:
period ⇒ <unset>
In the voice commands table, you can manage your commands with ease. Click the edit button to modify a command, or the delete button to remove it.
Sometimes a word may be mispronounced in various ways, or you might want the same action triggered by different phrases. Instead of creating separate commands for each variation, you can use the Alias feature. For example, if you say "ChatGPT," it might be interpreted as "check GPT" or "chat GPT." With aliases, create one command and add multiple pronunciation variants to it, ensuring the desired action is executed regardless of how it's spoken.
Here are some common use cases for custom voice commands. You can find more examples for various work domains at
Voice In might misinterpret spoken words due to accents, inability to disambiguate, an uncommon word, or other factors.
For example, a user who works for a company named iPacket reported that when she dictated her company's name, it typed it as "eye packet". To fix this, she can add a custom voice command eye packet => iPacket This command instructs the software to replace "eye packet" with "iPacket" whenever it encounters the word.
Another example: a user said his child's name "Sathyam" always gets dictated as "Satyam". To fix this, he can add a custom voice command "Satyam" => "Sathyam". This command instructs the software to replace "satyam" with "Sathyam" whenever it encounters the word.
Two more examples:
1. The user wants the word god always capitalized. To solve this, they can add a voice command "god" => "God". Now the word will be always be capitalized.
2. A user wanted "colon" to be inserted as ":". To fix this, they can add a voice command "colon" => ":".
By default, the browser's speech-to-text engine will censor curse words. You can use voice commands to reverse this. For example, you can add a custom voice command
s*** => suck
to remove censorship of that word.
Refer the article linked below for more information.
For example, a salesperson would add a voice command
thank you message
Thank you for purchasing from XYZ Store. Thanks, ABC
A doctor who is doing a general checkup of a patient would add
checkup text
General/Constitutional: overall health: unchanged Chills: denies Fever: denies Night sweats: denies
Respiratory: Cough: denies Hemoptysis: denies Shortness of breath: denies Wheezing: denies
This saves time with having to copy-paste this text block before every consultation.
Many German users (mainly users who use Swiss German) would want to use ß instead of ss in all dictated text. To achieve this, add a custom voice command mapping ß to ss. Set the match within a word flag to true.
\
Use the Export button to export a copy of your voice commands in CSV format. A backup of your custom voice commands can also be found . Currently, voice commands don't sync across your machines. If you are on a new machine or if you reinstall the Chrome extension, you may have to initialize the voice commands manually. We are working on automating this process.
For more details, check the article.
Use this action command to change the dictation language. Example: use spanish => <lang:es-ES> use english => <lang:en-US> The command requires the language code of the language. In the example above, en-US in the language code for English (United States). You can look up the language code for your language in .
Note that voice commands for delete last <2-20> words are already built-in for English. See the .
You can find the list of built-in voice commands at .
See the section above for instructions on creating a voice command.
[NEW] Voice In now has a built-in feature to disable the profanity filter for six languages. To enable this, go to the , scroll down to the Dictation Text Transforms section, and enable the "Disable profanity/adult words filter" feature.
If you have questions about custom voice commands, post on the .
