Major companies like Microsoft, Apple, and Meta, among others, are collaborating with the University of Illinois under the Speech Accessibility Project to improve speech recognition facilities.
Voice recognition RSS
You can now change your voice into Morgan Freeman, a pilot, an astronaut, and several others using this new feature by Voicemod. It is currently accessible on the beta version for PC users.
The latest iteration of the Common Voice Dataset was released today bringing with it over 13,000 hours' worth of crowd-sourced speech data, over 182,000 unique voices, and a total of 76 languages.
Researchers at MIT have created a framework—TextFooler—that brought down the prediction accuracy of certain NLP models from 90% down to under 20% by simply using synonyms in place of certain words.
Microsoft has patented an artificial speech generating device for communication sessions. The creation of such a device would help significantly improve the quality of audio communications.
A recently published patent by Microsoft hints towards the possibility of native voice input through digital assistants being integrated with third-party applications in the future.
A family from Portland, Oregon had a private conversation recorded and sent to one of the family's contacts through one of Amazon's Echo devices. Amazon has confirmed the "extremely rare" occurrence.
Amazon announced a slew of new features for third-party developers to leverage at its re:Invent conference in Las Vegas, including notification support and app personalization based on who's asking.
_medium.jpg)
Along with some new voice recognition features introduced by Barclays, customers can now make money transfers using Siri and Apple touch ID, without having to open the Barclays banking app.
Google today announced voice support for an additional 21 languages for its search, Gboard on Android and the Cloud Search API, adding Georgian, Swahili, eight Indian languages, among others.
Amazon has published detailed instructions on Github about how to make a Raspberry Pi-based DIY version of the Echo device - a product which uses Amazon's Alexa Voice Service.
Last September, Google added 'voice typing' to Docs. Starting today, you can edit using your voice to copy and paste, add bullets, select and replace text, and change styles.
Google Apps have been updated with a number of new features, so you "you can now spend less time trying to decipher your data, and more time making a point.”, as the company states.
Google has failed to reach an agreement to release a patch, leaving users vulnerable to hackers. The exploit allows spies to continue recording conversations through Chrome's voice recognition app.
Apple have revealed conservations you have with voice assistant Siri will be kept for up to two years. The announcement comes just days after questions were raised over users' privacy rights.
Apple are pushing additional changes to their voice assistant Siri in order to remove some exploits allowing Siri to be used for sexual or generally tasteless messages.
An iPhone hacker announced his success at porting Siri from iPhone 4S to the older iPhone 4 and even the iPod touch, bypassing Apple's preventative measures in doing so.
