Use Voice Recognition to fill out forms and dictate email with speech to text. Control the Internet with custom voice commands!
Speech Recognition Anywhere
With "Speech Recognition Anywhere" you can control the Internet with your voice. Use voice recognition to fill out forms and documents on the web! Dictate emails with speech to text!
No need to copy and paste your speech into a form input field. The speech you speak is automatically typed into any standard form on a web page hands free. You can also browse the web and completely control a website with voice commands. It can be used like a virtual assistant.
NEW! Speech Recognition Anywhere now includes text to speech, custom voice commands and scripting. See seabreezecomputers.com/speech for more information.
* Choose between dozens of languages and dialects for speech recognition
* Dictate emails and online documents
* Fill in forms with your voice
* Go to the next or previous field with your voice
* Go to any web page with your voice
* Switch tabs and navigate webpages with your voice
* Scroll page up or down
* Click on links and buttons with your voice
* Cut, Copy, Paste, Clear, Highlight
* Say "Show labels" to see labels to buttons on a webpage
* Say "Play (name of artist or song)" to play music instantly
* Text to Speech (tts)
* Three custom voice commands
* Use Voice Activation Wake Up Phrase: (Examples: 'Ok Google', 'Computer', 'Alexa', 'Hey Siri', 'Hey Cortana')
* Start "Speech Recognition Anywhere" when Chrome starts (Always Listening)
* Start "Speech Recognition Anywhere" in the background
* Submit search fields automatically
* Pause "Speech Recognition Anywhere" if audio is playing in a tab
* Virtual Assistant Mode
* Prevent system from sleeping while "Speech Recognition Anywhere" is listening
* Prevent screen from sleeping while "Speech Recognition Anywhere" is listening
* Unlimited custom voice commands
* Auto Punctuation
* Export/Import custom commands
NOTE: Google does not allow extensions to work on special chrome:// pages or the Chrome Web Store. Therefore, you will not be able to write a review directly at the Chrome Web Store using your voice. This is a restriction imposed by Google, not a limitation of the extension.
TIP: If speech recognition is not working on a specific website then you can try (1) refresh the web page or (2) restart your computer. This will often fix the problem.
Disclaimer of Warranty: Sea Breeze Computers makes no warranties as to the performance of this product or service. This extension uses Google's Web Speech API for speech recognition therefore we cannot guarantee the quality or the continued operation of the speech recognition service.
3/10/2021 - Version 1.4.5 - Bug Fix: Settings were not being saved after closing and reopening the Speech Recognition Anywhere tab.
3/9/2021 - Version 1.4.4 - Fixed bug in license status check. Now displays a notice if the user did not allow the extension to use the user's email address to get license information from Google's servers. Fixed "New Paragraph" to do two carriage returns. Changed speech to use insertText instead of insertHTML because Chrome now preserves undo history with insertText. But this will cause some user's custom commands to print HTML instead of inserting HTML. So there is now an insertHTML() custom command. Also changed replace_word() custom command to be able to remember the first match with $1 instead of $2 by using ?:
2/10/2021 - Version 1.4.3 - Added more error checking in the license status check. Changed title link to link to correct editions. Added more speech recognition languages such as Chinese Cantonese (Traditional).
1/30/2021 - Version 1.4.2 - Added "Disable Auto Gain Control" in Settings because Chrome changes the system microphone volume up and down on some computer systems. This setting will disable that from happening. Also fixed some more bugs with the license status. Check button now checks Chrome Web Store license and new payment processor license.
1/21/2021 - Version 1.4.0 - Added Dark Mode support and a Chrome keyboard shortcuts link. Added permissions "identity.email" to manifest.json because Google is deprecating Chrome Web Store payments so now developers of Chrome Extensions need to keep track of user email addresses to process license status.
10/23/2020 - Version 1.3.9 - Fixed the "Something went wrong" error with Twitter if you tried to do speech to text without typing at least one character first. Now you don't have to type on the keyboard first! Also, fixed bug caused by version 1.3.8. It caused some iframes to type every speech to text statement twice, such as evernote. Now it determines if it already typed the speech in the box.
10/13/2020 - Version 1.3.8 - Some websites with comment boxes in an iFrame were not inputting spoken speech. This seems to have been fixed. Also fixed a bug where some single word commands were being activated with a longer version of the word. For instance, endless and homeless were matching end and home for moving the cursor. This only happened if just one word was spoken but it has been fixed.
9/19/2020 - Version 1.3.7 - If a user put an invalid regular expression in a custom command then the speech recognition may stop working and only display an error in the developer tools console. Now it will display the regular expression error in the Speech Recognition Anywhere tab and in a tooltip. Also, on some pages the page would try to scroll up to the top to the yellow speech tooltip on speech being spoken. It seemed to only happen on Reddit. Fixed.
7/30/2020 - Version 1.3.5 - Some configurations of tinyMCE Editor were inputting speech twice into editor. This has been fixed. Some network errors and errors caused by conflicting extensions were causing the Speech Recognition Anywhere tab to become active. Added code to try to stop this. Also added suggested default activation hotkey: ALT+S. This will only work on first install if the hotkey is not already in use. Google Chrome's built in web speech API no longer seems to be processing "new paragraph" or "new line" so these have been added as built-in commands for Speech Recognition Anywhere. Also programmed it so that if a user needs to use a built-in phrase for a command they can create a custom command for that phrase. For example: Phrase: new line
Action: replace_word(new line)
This will allow sentences like, "I tried a new line of dog food" instead of entering a new line in the document.
2/24/2020 - Version 1.3.3 - Text-To-Speech (TTS) can now read in iframes as long as the origin is the same as the top page. Some features are not available in iframes such as specifying to "Read links" or "Read buttons" in iframes. It will be skipped. But "Read page" or "Read article" should read an iframe that is in the same origin if it is on screen. It will not select the text in the iframe as it is being read but it will put an outline around it for a second if "Highlight text being read" is checked in Settings. Also ESC key to pause/resume TTS now works on SRA tab as well as other tabs. Also fixed a bug with "Find phrase" where it was sometimes finding the phrase in the tooltip.
12/17/2019 - Version 1.3.2 - Fixed some bugs with using a voice command to click on form elements Select and Option when. Now changes the size to show the entire length of options instead of just 5. Fixed bug where it would not reselect the top option but would just highlight the select box. Also fixed a bug where it would sometimes not save the tts voice selected but revert back to the default Microsoft voice. It will now allow tts rate to go to a max of 10 for Microsoft voices. But Google voices only allow a max of 2. In Chrome 79 the developers added tab.pendingUrl which caused tab.url to no longer work so that when a user clicked on the Speech Recognition Anywhere icon again it would open a new tab with Speech Recognition Anywhere even though it was already open. Added code to fix this behavior.
5/29/2019 - Version 1.3.0 - The converter can now do advanced scripting with the script() command or by changing semicolons in code to url encoded %3B . Also saying "Show labels" now labels text inputs and textareas.
5/6/2019 - Version 1.2.9 - A custom command with replace_word() can now have multiple phrases separated by a vertical pipe | . Also fixed a bug where "Delete last word" was deleting the entire last phrase instead.
3/25/2019 - Version 1.2.7 - If the phrase "Play|Listen To|Youtube" is spoken then the video will no longer be played in the current tab unless the current tab is youtube. Also if the "Action" field is empty for a Custom Command then that command will not be acted upon. Also switching tabs by url or keyword in title is now easier and better.
3/7/2019 - Version 1.2.6 - New simple word replace feature. For example, to replace "ok" with "okay" in your speech then use this custom command:
Fixed custom commands bug where a phrase separated with vertical pipes like ^hi|hello$ will match "hi there" and "well hello" without wanting it to.
1/20/2019 - Version 1.2.3 - Improved Screen Reader features such as reading of ARIA roles and html tag names for certain items like IMG and Hyperlink. Also a "Simple Mode" that does not read ARIA roles and html tag names.
1/8/2019 - Version 1.2.1 - Fixed a bug where "Read selected text" command was not working. Also added to "Read notification" command to also read role="dialog" and role="alertdialog".
1/5/2019 - Version 1.2.0 - The custom command speak() for Text-To-Speech was no longer working in Chrome 71 because Chrome is trying to stop audio from playing automatically without user interaction. Changed the Text-To-Speech speechSynthesis to play in the background script instead of the page. Also added Screen Reader features. Now you can say "Read selected|highlighted|article|website|site|links|page|webpage|paragraph|alert|notification|status". Press ESC to stop reading or say "Stop reading".
12/28/2018 - Version 1.1.9 - Translated commands to Italian.
12/15/2018 - Version 1.1.8 - Now there is an option in Settings to "Remove Google's Auto Capitalization" where Google's speech recognition sometimes adds capitalization to phrases that are the same as sport's teams, movie titles or song titles, etc.
Also accidentally removed "FREE TRIAL (x days left)" in version 1.1.3. Fixed.
10/8/2018 - Version 1.1.6 - Better error handling in multiple languages. Faster Initializing. Added Swedish punctuation and commands.
9/30/2018 - Version 1.1.5 - There is now a delete button for custom commands. Also added punctuation for the Czech language.
9/21/2018 - Version 1.1.4 - Improved language support including French, Japanese, Chinese Simplified and Chinese Traditional.
9/19/2018 - Version 1.1.3 - Improved language support including Italian.
6/5/2018 - Version 1.0.9 - Speech Recognition Anywhere had stopped working in Google Docs because of a Google Docs update. This has been fixed. Also in Google Docs it was capitalizing the first letter after every speech recognition phrase even if it was not a new sentence. This has been fixed as well.
4/8/2018 - Version 1.0.5 - Export/Import custom commands.
3/29/2018 - Version 1.0.4 - Custom command word replace now appears in Speech Recognition tab text box. Also previously if "Phrase" or "Action" fields for custom commands contained a space for the first character then it would not be recognized. This has been fixed by removing the extra space.
Also Speech Recognition Anywhere now works better with Facebook Messenger without having to press Spacebar first. And to send the message you can say "Press Enter" or "Click Send" at messenger.com.
Also works better with Google Hangouts at hangouts.google.com. You can now say "Press Enter" to send the text.
Also new and improved custom command: keypress(key, ctrl, alt, shift) that now works with more websites.
3/3/2018 - Version 1.0.1 - New Feature: Speech Recognition Auto Punctuation. With Auto Punctuation enabled you no longer need to say "period" or "question mark" at the end of a sentence. Just pause briefly after each sentence and the punctuation will be added automatically!
Speech Recognition Anywhere now works in Word Online with a few bugs: It first tries to type the speech in "Tell me what you want to do" box and you have to click in the word processor box again. "New Line" and "New Paragraph" commands do not seem to work.
Also improved speech interaction with HTML FORM SELECT fields. Now saying "Click on select" or "Click on (an option)" will open the select menu bigger and fire a Change event if the selection is changed.
Bug Fix for TinyMCE: Now it capitalizes the first letter of a sentence if it follows punctuation.
Bug Fix for CKEditor: After saying "new line" now the first letter of a sentence is capitalized.
2/1/2018 - Version 0.99.9 - All commands have been translated to work in Spanish!
Also all the text in the action field command print_text(text) now preserves the letter case as it is written by converting all letters to unicode html entities. Previously, words at the beginning of sentences were auto-capitalized.
11/20/2017 - Version 0.99.4 - Added punctuation and commands in German language (Deutsche).
9/28/2017 - Version 0.99.2 - Now Speech Recognition Anywhere works with Google Docs! Also fixed a bug where it would try to scroll to a cursor in a textarea but sometimes scroll the screen when it did not need to.
9/11/2017 - Version 0.99.0 - Added punctuation for Spanish and Portuguese. Also added locales for Spanish and Portuguese. Fixed bug where the page would scroll past the textarea, such as youtube comments, when trying to scroll to the cursor. Fixed by adding elRect.height > 0. Added "Find word or phrase" command. Fixed bug where "new tab" would not use Speech Recognition because it was using "chrome://newtab/". Changed to "https://www.google.com/_/chrome/newtab" and now it works again.
2/1/2017 - Version 0.98.4 - Added Text to Speech (tts) and scripting. Now you can create custom commands that talk back to you or read text in an element in a web page. In the "Action" field type in: say(Hello There) . Or to go to a webpage and speak the text in an element, type in the "Action" field: http://website/com;speak(element_id) . For more information go to seabreezecomputers.com/speech/ . Also changed recognition.continuous to false instead of true to try and solve slowness with the recognition result or multiple second delay with the Web Speech API that some users have.
10/28/2016 - Version 0.98.1 - Languages and dialects are here! Added an option in settings to choose between dozens of languages and dialects for speech recognition including: Afrikaans, Arabic, Basque, Bulgarian, Catalan, Czech, Dutch, English, Filipino, Finnish, French, Galician, German, Greek, Hebrew, Hungarian, Icelandic, Italian, Indonesian, Japanese, Korean, Latin, Lithuanian, Chinese, Malaysian, Norwegian, Persian, Polish, Portuguese, Romanian, Russian, Serbian, Slovak, Spanish, Swedish, Turkish, Vietnamese, Zulu. We need volunteers to translate punctuation and commands into these languages from English. Please contact firstname.lastname@example.org
09/12/2016 - Version 0.9 - Speech Recognition Anywhere first release.