1 844.742.7448

Email Us

Remote Support

Contact

Posts tagged "gets"

  1. Home
  2. Posts tagged gets

Windows 11 speech recognition feature gets ditched in September 2024 – but only because there’s something better
  • No Comments

Windows 11’s voice functionality is being fully switched over to the new Voice Access feature later this year, and we now have a date for when the old system – Windows Speech Recognition (WSR) – will be officially ditched from the OS.

The date for the replacement of WSR by Voice Access has been announced as September 2024 in a Microsoft support document (as Windows Latest noticed). Note that the change will be ‘starting’ in that month, so will take further time to roll out to all Windows 11 PCs.

However, there’s a wrinkle here, in that this is the case for Windows 11 22H2 and 23H2 users, which means those still on Windows 11 21H2 – the original version of the OS – won’t have WSR removed from their system.

Windows 10 users will still have WSR, of course, as Voice Access is a Windows 11-only feature.

Analysis: WSR to go MIA, but it’s A-OK (for the most part)

This move is no surprise as Microsoft removed Windows Speech Recognition from Windows 11 preview builds back at the end of 2023. So, this change was always going to come through for release versions of Windows 11, it was just a question of when – and now we know.

Will the jettisoning of WSR mean this feature is missed by Windows 11 users? Well, no, not really, because its replacement, Voice Access, is so much better in pretty much every respect. It is leaps and bounds ahead of WSR, in fact, with useful new features being added all the time – such as the ability to concoct your own customized voice shortcuts (a real timesaver).

In that respect, there’s no real need to worry about the transition from WSR to Voice Access – the only potential thorny issue comes with language support. WSR offers a whole lot more in this respect, because it has been around a long time.

However, Voice Access is getting more languages added in the Moment 5 update. And in six months’ time, when WSR is officially canned (or that process begins), we’ll probably have Windows 11 24H2 rolling out, or it’ll be imminent, and we’d expect Voice Access to have its language roster even more filled out at the point.

Those on Windows 11 21H2 will be able to stick with WSR as observed, but then there’s only a very small niche of users left on that OS, as Microsoft has been rolling out an automatic forced upgrade for 21H2 for some time now. (Indeed, this is now happening for 22H2 as of a few weeks ago). Barely anyone should remain on 21H2 at this point, we’d imagine, and those who are might be stuck there due to a Windows update bug, or oversight during the automated rollout.

Windows 10 users will continue with WSR as it’s their only option, but as a deprecated feature, it won’t receive any further work or upgrades going forward. That’s another good reason why Windows 11 users should want to upgrade to Voice Access which is being actively developed at quite some pace.

You might also like…

TechRadar – All the latest technology news

Read More

ChatGPT gets a big new rival as Anthropic claims its Claude 3 AIs beat it
  • No Comments

AI company Anthropic is previewing its new “family” of Claude 3 models it claims can outperform Google’s Gemini and OpenAI’s ChatGPT across multiple benchmarks.

This group consists of three AIs with varying degrees of “capability”. You have Claude 3 Haiku down at the bottom, followed by Claude 3 Sonnet, and then there’s Claude 3 Opus as the top dog. Anthropic claims the trio delivers “powerful performance” across the board due to their multimodality, improved level of accuracy, better understanding of context, and speed. What’s also notable about the trio is they’ll be more willing to answer tough questions. 

Anthropic explains older versions of Claude would sometimes refuse to answer prompts that pushed the boundaries of the safety guardrails. Now, the Claude 3 family will have a more nuanced approach with its responses allowing them to answer those tricky questions.

Despite the all-around performance boost, much of the announcement is focused on Opus as being the best in all of these areas. They go so far as to say the model “exhibits near-human levels of comprehension… [for] complex tasks”.

Specialized AIs

To test it, Anthropic put Opus through a “Needle In a Haystack” or NIAH evaluation to see how well it’s able to recall data. As it turns out, it’s pretty good since the AI could remember information with almost perfect detail. The company goes on to claim that Opus is quite the smart cookie able to solve math problems, generate computer code, and display better reasoning than GPT-4. 

The technology isn’t without its quirks. Even though Anthropic states their AIs have improved accuracy, there is still the problem of hallucinations. The responses the models churn out may contain wrong information, although they are greatly reduced compared to Claude 2.1. Plus, Opus is a little slow when it comes to answering a question with speeds comparable to Claude 2.

Of course, this isn’t to say Haiku or Sonnet are lesser than Opus as they have specific use cases. Haiku, for example, is great at giving quick replies and grabbing information “from unstructured data”. Also, it’s not as good at answering math questions as Opus. Sonnet is a larger-scale model meant to help people save time at menial tasks and even parse lines of “text from images”, while Opus is ideal for large-scale operations.

Changing the internet

Both Sonnet and Opus are currently available for purchase although there is a free version of Claude on the company website. A launch date was not given for Haiku, but Anthropic states it’ll be released soon. 

As you can probably guess, the Claude 3 trio is meant more for businesses looking to automate certain workloads. Your experience with the group will likely come in the form of an online chatbot. Amazon recently announced it’s going to be implementing Anthropic’s new AIs into AWS (Amazon Web Services) giving websites on the platform a way to create a customized Claude 3 model to suit the needs of brands and their customers.

If you're looking for a model suited for everyday use, check out TechRadar's list of the best AI content generators for 2024.

You might also like

TechRadar – All the latest technology news

Read More

Copilot gets a big redesign and a new way to edit your AI-generated images
  • No Comments

It’s been one year since Bing Chat received its generative AI power-up and we’ve seen it change a lot since including a rebranding into Copilot. To celebrate the first anniversary, Microsoft decided to redesign Copilot’s homepage as well as introduce a new editing feature.

The company states when you visit the AI’s engine desktop website, “you will see… a cleaner, sleeker look”. In the middle of the page is a revolving carousel of sample prompts with an accompanying image. Its purpose, according to Microsoft, is to give you an idea of what Copilot is capable of; to get those creative juices flowing. It is certainly more engaging than the previous version. The old page had three sample text prompts next to each other with no indication that it could create images.

Copilot on mobile is receiving an identical update. The app has the same carousel of sample prompts with a picture above to give you some ideas. You also have the option to toggle GPT-4 for better results. Activating it turns the software’s blue accents to purple. 

Tweaking prompts

As for the feature mentioned earlier, it’s called Designer. It allows you to make tweaks to generated content like highlighting certain aspects, blurring the background, or adding a unique filter. As an example, let’s take Copilot’s suggestion of creating an image of an animal wearing a football helmet. Moving your cursor over the picture makes a bold white line appear around an object. Clicking it highlights the portion. 

A couple of options appear at the bottom of the window. We chose to tell Copilot to make the colors pop. After a few seconds, the finished product appears. You can then either undo the effect or keep it. For filters, you have eight to choose from. Pixel art, block print, and claymation are some of the selections. Like the edits before, applying a filter takes a few seconds. 

Image 1 of 3

Copilot highlighting subject

(Image credit: Future)
Image 2 of 3

Copilot making colors pop

(Image credit: Future)
Image 3 of 3

Copilot pixel filter

(Image credit: Future)

Designer is free for everyone to try out. However, subscribers to Copilot Pro will be given extra tools. They can resize generated content and regenerate images into either a square or landscape orientation. Microsoft says it eventually roll out a “Designer GPT” to Copilot. The company calls it a canvas of sorts where people can “visualize [their] ideas.” If we had to take a guess, it could be a publicly available GPT model that you can use to create editing tools. OpenAI offers a similar service with its online store. We reached out to Microsoft for more details. This story will be updated at a later time.

Check out TechRadar's list of the best free drawing software for 2024 if you'd like to find a way to make the edits yourself.

You might also like

TechRadar – All the latest technology news

Read More

Google Bard finally gets a free AI image generator – here’s how to try it
  • No Comments

Considering how popular image generating is right now, Google Bard’s recent update that adds AI image generation seems all but inevitable.

According to the official Google Bard update page, using the tool is quite simple. You can activate the AI image generator by simply entering “a few words” into the search bar, “starting with English prompts.” Then you “click ‘Generate more’ for more options and download the ones you like.”

Those generated images are stored in pinned chats, recent chats, and Bard Activity, and can be deleted from your Bard Activity by deleting the prompt that generated them.

Bard’s image generation is powered by its updated Imagen 2 model, which is meant to balance speed with quality to deliver photorealistic images. The new feature is available worldwide including in the US, except in the following regions/countries: European Economic Area (EEA), Switzerland, and the UK. It’s also only available in English and only to those 18 and above, though how Google would enforce the age restriction beyond a generic ‘Yes’ or ‘No’ question is unclear.

In terms of responsibility, Google states in its official blog that “Bard uses SynthID to embed digitally identifiable watermarks into the pixels of generated images,” which makes its AI-generated images distinct from works created by humans. Most likely this feature was developed to prevent people from using generated images for commercial use, which tracks since that would open up a legal can of worms.

Google also asserts that it seeks to limit violent, offensive, or sexual content from the training data, as well as applying filters to prevent named people from being involved in image generation. 

Bard with Gemini Pro will also be enhanced with this new update. The update page states that “Bard will be far more capable at things like understanding, summarising, reasoning, brainstorming, writing, and planning.” This upgrade to Bard’s AI is most likely what allowed Google to offer the free AI image generator tool in the first place, along with the support.

Google Bard Image Generation

We created the first two images with a simple prompt and clicked “Generate more” to see the second two. (Image credit: Future)

Google Bard is taking over 

Google has been going all in with its Bard AI, even using it to slowly replace Google Assistant, which was the tech giant’s previous answer to Apple’s Siri. It was discovered that Google changed the greeting pop-up for various devices from ‘I’m Assistant with Bard’ to simply ‘I’m Bard.’

It was also announced that Google would be removing 17 features from Assistant in the coming weeks, including playing audiobooks on Google Play Books through voice command and asking for information about your contacts. The official announcement even implied that more features would be removed.

You might also like

TechRadar – All the latest technology news

Read More

Apple Vision Pro gets new mixed-reality weather forecasts as we wait for more apps
  • No Comments

Apple’s Vision Pro is just a couple of days away from launching, and while Apple has highlighted some astonishing apps that are coming to the device, there have also been questions over how many apps will be available for users upon release. Well, we now know that one of the best iPhone apps will be among that number, and it could provide you with a great way to use Apple’s mixed-reality headset.

The app is the ever-popular Carrot Weather. As the name suggests, this app provides you with forecasts, weather warnings, and everything in between. It’s known for its caustic sense of humor – your reports are delivered by a robot whose outlook ranges from gentle to misanthropic – and that’s helped it stand out among a slate of other excellent weather apps.

On the Vision Pro, though, Carrot Weather is going to do something a little different. According to screenshots posted by the M1 Astra account on X (formerly Twitter), Carrot Weather will come with a floating planet Earth that shows the weather at whatever global location you select, as well as mini-games you can play via an tool at the bottom of the app window.

That comes in addition to the range of weather data you’d expect from an app like this, including temperatures, wind speeds, sunrise and sunset times, and more. That information will be displayed in a large floating window that shows far more data at a glance than the Carrot Weather iPhone app is able to do.

As spotted by MacRumors, Carrot Weather developer Brian Mueller said the floating globe would be the “marquee feature” of the app and that “it's just really cool being able to look at a globe floating in your living room.” It shows one of the ways the Vision Pro may let apps do things they never could on an iPhone or iPad.

Is the Vision Pro selling well?

The Carrot Weather app shown in augmented reality using Apple's Vision Pro headset.

(Image credit: Grailr)

It comes as news emerged that Apple has reportedly sold around 200,000 Vision Pro headsets in the first 10 days since pre-orders opened on January 19 (via MacRumors). Given pre-release predictions, that seems like an impressively high figure, but there is a caveat to those numbers.

Analysts had previously predicted that Apple would sell about 400,000 headsets in the entire first year of its availability. Apple tipster Ming-Chi Kuo, meanwhile, forecast that Apple would produce between 60,000 and 80,000 units for the device’s launch, suggesting that even Apple was unprepared for the level of demand.

That said, Kuo also claimed in a later report that Apple sold between 160,000 and 180,000 headsets during the pre-order weekend. That suggests that demand might have slowed in the days since then, as Apple has now apparently hit 200,000 units sold after 10 days.

The big question is whether Apple can sustain Vision Pro demand throughout the year. A large number of those 200,000 sales would likely have come from developers, hardcore Apple fans and early adopters. Will Apple be able to sustain this level of sales going forward, or will interest slowly die off over time? We’ll be watching closely.

You might also like

TechRadar – All the latest technology news

Read More

Shazam finally gets the upgrade that all headphones lovers have been waiting for
  • No Comments

Shazam's latest update added a feature you may be surprised to learn wasn’t already present: the ability to identify music through your headphones.

The way it works, according to the patch notes on the App Store, is you first launch the app whenever you find a song on YouTube, TikTok, or wherever. Tap the headphone icon on-screen to tell Shazam a pair of headphones are connected and to begin listening. From here, you can go to the video with the mysterious song and the app will pick it up – hopefully. 9To5Mac states the tool works on “almost any… app you can think of”, alluding to the possibility that Shazam may have some blind spots. Once you’re done, you can find the name of the song on the app’s interface.

It doesn’t matter what kind of headphones you use. The patch notes confirm the feature will work through wired and wireless devices alike. Nor does there appear to be any hardware restrictions either. You won't need to own a pair of AirPods to gain access.  It works with third-party brands.

Additionally, Shazam will continue to work normally even if a pair of headphones are connected. The feature won’t interrupt or prevent the app from listening to songs playing in your immediate vicinity.

We should also mention that the headphone listening upgrade is also available to Android users. The app’s listing page on the Google Play Store has been updated to now include a line mentioning the new functionality.

Possible macOS expansion

It’s not a major upgrade as you can see, but it does introduce a new level of utility to the app. You should now have an easier time finding out the names of songs you stumble across on social media.

It’s unknown if this function will roll out to Shazam on macOS. According to the Mac App Store listing, the last time Shazam on Mac saw a patch was August 11, 2022, so it’s been a while. We think this version is overdue for an update. So we reached out to Apple asking if there are plans to expand the headphone listening feature to its other rendition of Shazam. This story will be updated at a later time.

Until then, check out TechRadar’s list of the best headphones for 2024.

You might also like

TechRadar – All the latest technology news

Read More

Windows 10 gets a fresh lock screen weather widget and bug fixes in new update
  • No Comments

Microsoft has dropped a new preview update (KB5034203), bringing an updated lock screen weather widget to Windows 10 alongside regular bug fixes. The update also comes with changes to comply with the European Union’s Digital Markets Act – however, these changes are set to roll out gradually rather than all at once.

The update will bring more life to the weather experience on Windows 10, including more interactive weather updates, allowing you to hover your cursor over the widget on your lock screen to see more information about current weather conditions in your area. Clicking on the weather ‘card’ will then sign you into your device and open up Microsoft Edge with a full weather broadcast.

Alongside the updated widget, update KB5034203 comes with other fixes such as addressing issues that affect BitLocker data-only encryption, the failure of some apps in Windows 10 to respond to changed keyboard language, and a fix for an annoying bug that caused the Internet Explorer shortcut to become unresponsive for some users.

Optional, but still worth an update

In terms of compliance with the European Union’s Digital Markets Act, Microsoft released a blog post directed to Windows Insiders back in November addressing plans to ensure Windows 11 and Windows 10 follow the set obligations.

The Digital Markets Act includes a wide range of requirements that Microsoft will have to adhere to within the EU – most notably, the ability for users to uninstall any Windows app, including Cortana and Edge. According to the blog post, Microsoft will “be updating Windows 10, version 22H2 and Windows 11, version 23H2 PCs in the EEA to be compliant by March 6th, 2024.”

So, we can expect more of these smaller-scale updates in the coming months to both Windows 10 and Windows 11. I'm pleased to see Microsoft continuing the effort to keep Windows 10 updated, considering many people are still quite hesitant to jump ship to Windows 11.

KB5034203 is an optional preview update at the moment, so you’ll have to install it manually if you’d like to check these features out yourself right now. Head over to the ‘Update and Security’ section of your setting and click the ‘Windows Update’ tab. There you’ll see a list of optional updates available and a link to download and install the update. Enjoy!

You might also like…

TechRadar – All the latest technology news

Read More

Microsoft forges ahead with adding new features to Windows 10 as the OS gets another Windows 11 capability
  • No Comments

Remember when Microsoft said Windows 10 wouldn’t get any more features, and was effectively on ice, with the exception of minor tweaks? Well, the reversal of that policy is clear enough now, as another inbound feature for Windows 11 is also coming to its predecessor OS.

Namely a ‘richer weather experience’ – including dynamic, interactive weather updates – on the lock screen.

This is something that’s already been spotted in testing for Windows 11, and is now in the latest preview build for Windows 10 (version 19045.3992).

It means that when you hover over the weather on the lock screen, further info is revealed. On top of that, if you click the weather card, you get the MSN website weather forecast popping up (in Microsoft’s Edge browser, as you might guess).

This will be the default setup for the lock screen if its status is set to ‘none’ Microsoft explains, but you will be able to turn it off if you wish.

Analysis: Thawing the feature freeze

Okay, so this is no big deal as a feature, but nonetheless it is a new piece of functionality, and not a minor tweak. Of course, the other addition since Microsoft announced that feature freeze for Windows 10 was the introduction of Copilot – and you couldn’t get a more major move than that.

As Windows Central (Zac Bowden), which spotted this latest development, points out, late last year, when Copilot was announced for Windows 10, Microsoft did say it was revisiting its decision to halt features for the older OS, and planned to make ‘additional investments’ in the operating system. It appears that this is happening, then, and we can likely expect further features to be bolted on to Windows 10 as 2024 rolls onwards.

Furthermore, Bowden has heard on the grapevine that Microsoft will be reopening the Windows 10 Beta channel for testers, which is a clear suggestion that more features will be on the way for the operating system – as they’ll need to go through a full testing process.

You might also like…

TechRadar – All the latest technology news

Read More

Windows 10 gets security boost and bug fixes in Microsoft’s first big update of 2024
  • No Comments

Microsoft might be pushing forward with integrating AI into as many aspects of Windows 11 as possible, but it’s not totally forgotten about Windows 10 users. The older version of Windows continues to be very popular among Windows’ user base, and fortunately for them, Microsoft has just released update KB5034122 for Windows 10 that brings an array of bug fixes and serious security upgrades. 

Two of the bugs that the update addresses are to do with smart card usage and an issue with scroll bars. Maybe not the most thrilling updates, but this is pretty in line with Microsoft’s messaging about Windows 10. 

According to the tech titan, it’s more or less closed up shop when it comes to working on significant new features for Windows 10 and users shouldn’t expect to see any major changes in the future. Update KB5034122 serves as evidence of this with it being mostly maintenance and fixes from Microsoft, but let’s not forget that Microsoft’s shiny new all-in-one AI assistant, Windows Copilot, was made available to Windows 10 users last year. We’ll have to see if Copilot will see upgrades and improvements in Windows 10 considering that its current functionality is fairly limited.

Microsoft Teams copilot

(Image credit: Microsoft Teams)

What's new in update KB5034122

This update tackles security issues, as well as a quality upgrade to Windows 10’s servicing stack, the Windows component that enables users to install Windows updates. Microsoft also gives more details about the bug fixes that are included in this update: 

You can find a full rundown of what this update addresses on Microsoft’s Support blog, and it does make note of some known issues that still exist in this version of Windows 10 and gives suggested workarounds with instructions. It follows up each workaround for each presently-existing problem with the following statement to reassure Windows 10 users: 

We are working on a resolution and will provide an update in an upcoming release.

KB5034122 should be prompted for install on Windows 10 devices automatically because it’s a security update, but if for whatever reason your Windows 10 device has not downloaded it already, you can download it manually. You should definitely do this as it’s important to have the most up to date security fixes no matter what Windows version you use, and you can get it from the Microsoft Update Catalog. 

Good for Microsoft for keeping an eye on Windows 10 and recognizing that it remains a fan favorite. However, it’s clearly determined to get as much use out of its investment and collaboration with OpenAI, utilising GPT technology however it can.

Recently, Windows watchers have spotted that Notepad is getting a ChatGPT-powered writing assistant and text editing AI tool, with some users expressing that they’d rather Notepad stayed the simple, straightforward app that it came to be known as. Perhaps as Microsoft goes down the path of ramping up AI integration, Windows 10 will be a refuge option for those that want their operating system and apps to be a little less intelligent. 

YOU MIGHT ALSO LIKE…

TechRadar – All the latest technology news

Read More

ChatGPT gets its equivalent of the App Store – here are the best early GPTs
  • No Comments

OpenAI has officially launched its GPT Store, allowing a select group of users and official partners to share customized chatbots with the community.

The platform won't be available to the wider public, according to the company You must have a subscription to ChatGPT Plus ($ 20 / £16 a month), Enterprise, or the newly formed Teams plan. Once you buy the subscription, you’ll be given access to a variety of GPTs across multiple categories. These include writing, programming, and art generation – some will even provide lifestyle advice. One, in particular, can help you refine designs for a tattoo you’re thinking of getting. 

The GPT Store was originally announced this past November during the company’s first DevDay conference. It was shown alongside OpenAI’s then-new create-a-chatbot service. The store was supposed to open later that month but was delayed multiple times, most likely as a consequence of the sudden ousting and reinstatement of CEO Sam Altman.  

Curated chatbot

Although the average person can’t try out chatbots, you can head over to the store’s page right now to see what’s available. Make sure you’re not logged into your account: if you are, you’ll get a message telling you to subscribe to ChatGPT Plus. OpenAI will highlight four GPTs at the top of the page every week. At the time of writing, AllTrails, Consensus, Code Tutor, and Books are being recommended.

The first one, AllTrails, will suggest nature trails for you to explore. Consensus, formerly known as ResearchGPT, has access to 200 million academic papers and can answer your tough science questions. Code Tutor, developed by the education platform Khan Academy, will look over your recently made computer code and make suggestions on how to improve it. Finally, there’s Books, the most mysterious of the bunch. Its description is rather vague, however, if we had to guess what it does, we'd say it provides book recommendations.

GPT Store front page

(Image credit: Future)

Below that are a couple of other lists highlighting chatbots currently trending among the community and those made by OpenAI’s internal team. There are plans to implement a revenue program that will allow people to make money from their creations. Creators will “be paid based on user engagement” with their chatbots. Details are light at the moment. All that’s known currently is it’ll launch sometime in Q1 2024 in the United States.

Follow the rules

Anybody with a subscription can create a GPT. OpenAI states you don’t need any coding skills, although you will need to follow usage policies and brand guidelines. It wants to keep things clean. 

You can read the rules on the official website, but just to give you an idea, users cannot A) compromise people’s privacy, B) create an AI that may impair the well-being of others, or C) use the platform to spread misinformation. Break the rules, and the company will restrict your ability to share or monetize your work. If you run into any of these rogues on the GPT Store, OpenAI asks that you report them.

If you have a creation, you can share it by first saving your GPT and then selecting Everyone in the process. This gives all people on the platform access. Once done, you’ll need to “verify your Builder Profile” in the Settings menu. After you do all that, you should see your chatbot on the storefront.

While we have you check out TechRadar's list of seven tips for ChatGPT beginners. That generative AI can be a little tricky to use.

You might also like

TechRadar – All the latest technology news

Read More