transcription
Latest
Microsoft brings live captions and subtitles to PowerPoint
Microsoft is rolling out a new accessibility feature for PowerPoint, one that stands to help speakers ensure their presentations are understood by their entire audience. The company notes that in cases where audience members are hard of hearing or speak a different language than the presenter, the speaker's presentation may not be fully communicated, but a new tool aims to solve that problem. Soon, PowerPoint will be able to transcribe and caption what a speaker is saying in real-time, giving audience members the option of reading what a presenter is saying. And the tool will also be able to translate speech into different languages, subtitling the presentation into a language of the speaker's choice.
Google's call screening transcripts roll out to Pixel owners
You no longer have to stare at your Pixel to see whether or not that Assistant-screened call was worth answering. Pixel owners have reported that Call Screen transcripts are starting to reach their devices, including older devices as well as the Pixel 3. Check the recent calls in your phone app, choose Call Details and you'll find a See Transcript option if you have the feature. If you do, you can review what a mysterious caller said and decide whether or not it's worth a follow-up.
Microsoft OneDrive will use AI to make searchable video transcripts
You've probably had that moment where you wanted to track down an important piece of information from a video, but weren't sure when it was said. If so, Microsoft wants to come to your aid -- it's introducing media searching in OneDrive (and SharePoint, for that matter) that uses AI to transcribe audio and video. The feature will show you timestamped quotes alongside the media viewer itself, with a handy search box helping you track down that elusive phrase.
Google Voice integration with Sprint is coming to an end
Google Voice, the service that gives US users a free phone number they can use for calls, texts and messages on devices other than a smartphone, recently got a big overhaul. Left intact was Sprint's unique arrangement with Google that allowed its customers to use their Sprint phone number for Google Voice. Google has informed Sprint customers that the arrangement is about to end, however, thanks to changes coming to Sprint's network.
Google voice recognition could transcribe doctor visits
Doctors work long hours, and a disturbingly large part of that is documenting patient visits -- one study indicates that they spend 6 hours of an 11-hour day making sure their records are up to snuff. But how do you streamline that work without hiring an army of note takers? Google Brain and Stanford think voice recognition is the answer. They recently partnered on a study that used automatic speech recognition (similar to what you'd find in Google Assistant or Google Translate) to transcribe both doctors and patients during a session.
AI takes the headaches out of transcribing voice recordings
Ask many interviewers about their least favorite part of the job and they'll almost always point to transcription. It can take hours to turn even a short chat into text, which is a serious pain for everyone from reporters to police interrogators. China tech giant Baidu may have a smarter approach: artificial intelligence. It just released a beta for SwiftScribe, a transcription app that uses a neural network to make sense of speech. The software not only promises relatively accurate speech-to-text processing thanks to training on "thousands of hours" of recordings, but learns from edits. It should account more for how people actually speak, saving you from making a load of edits.
Google Voice gets a long-overdue visual refresh
Google Voice is undoubtably useful, but it has also been neglected for a long time. Fortunately, the rumors were true: Google just released a pretty major visual overhaul of Voice, and there are a handful of new features on board as well. Most obvious is that service now fits in with Google's Material Design language, something that rolled out to just about all of Google's apps a long, long time ago.
Google wants your help to improve its automatic translations
Google's ability to interpret and translate handwriting isn't perfect. Sometimes you'll scribble a word or take a photo of a restaurant menu on holiday, only to have a garbled mess thrown back at you. To help its "smart" assistants and services, Google has released a new app on the Play Store called Crowdsource. It's a bare-bones affair, asking you to transcribe digital squiggles and photographed road signs. There are no discernible rewards, only the occasional message ("you're great!") and meaningless 'milestone' when you've completed a certain number of tasks. In short, you'll need to really love Google to open the app more than once.
Apple reportedly wants to turn Siri into your receptionist
Apple is testing a service that will let Siri take your calls, record them and transcribe them to text, according to Business Insider. The company is reportedly referring to it as iCloud Voicemail, and it's similar to the existing visual voicemail service. However, instead of playing a pre-recorded message to your caller when you can't pick up, Siri will take over the chore. It can then let certain contacts know where you are and why you can't take the call, provided you give permission. The voice message will then be shunted over to Apple's servers and transcribed into text.
Google Voice transcriptions will soon actually make sense
One of the most prevalent qualms users have of Google Voice is its occasionally accurate (but usually absurd) interpretations of what's being said. However, with the upcoming public debut of the Project Fi cellular service, Google has reportedly greatly improved its transcription service. According to a post on the company's blog, Google's managed to reduce its transcription error rates by nearly 50 percent by leveraging a "long short-term memory deep recurrent neural network." Users don't even have to change their routine to take advantage of the new system, just keep using Voice and Fi as they always have. [Image Credit: shutterstock]
Google will finally add iPhone-like visual voicemail to Android
Android users on select networks will soon get native "visual voicemail," a feature that iPhone users have enjoyed since forever. In case you're wondering, that's a way of checking and deleting voice mails via an app, rather than having to call a carrier number and go through them one by one. The feature was spied by Android Police on a support ticket for the upcoming Android M release and via Google+ user Danny Hollis. Hollis showed a screen cap of the new interface (below), and said it's now implemented for T-Mobile in a preview build.
Google Voice plans to make transcriptions better with your help
What's the silliest Google Voice transcription you've gotten? That question might have come up during a meet-up with tech-loving friends before -- after all, you're not the only user who's ever received garbled voicemail-to-text messages. In fact, even Google Tech Lead Manager Alex Wiesen admits they can be "unintentionally hilarious," to the point that the company's now asking for your help to make transcriptions better. Now, when you log into Google Voice on the web, you'll be given the choice to share your voicemail messages (anonymously, thank goodness) to be analyzed for accuracy by automated systems. While you can already submit individual messages for analysis, you'll automatically be sending the system all your messages, all the time, if you decide to participate in this project. Don't worry, though: you can always disable it through Google Voice's settings page in case you change your mind later on.
Dragon adds Recorder app for time-shifted desktop dictation
It's all about speech on the iPhone 4S, from the systemwide dictation features to the inscrutable but very helpful Siri assistant. The fine folks at Nuance (suppliers of some of the underlying IP that powers the 4S voice savvy) have made a big move into the mobile space; the company already has a suite of iOS apps that cover several speech-related functions. There's Dragon Dictation for text entry, Dragon Go! and Search for finding what you need, Dragon for SalesForce to work with your CRM system, two Dragon Medical apps for search and recording, and the Dragon Remote Mic app that turns your iPhone or iPod touch into a networked microphone for the company's desktop dictation apps (Dragon Dictate for Mac or NaturallySpeaking for Windows). Now there's another member of this growing family: the Dragon Recorder app. Dragon Recorder is a free and straightforward voice recording app designed to pair with the company's transcription software; on the Mac, that means the $150 MacSpeech Scribe application. You can use Recorder to record your voice (only yours; Scribe and its Windows sister product are speaker-dependent) on the go, and then easily transfer the recordings via sync or Wi-Fi browser sharing for later transcription. Of course, you could use the built-in Voice Memos application to achieve much the same result, but you wouldn't get the Wi-Fi sharing feature. Then again, if you're planning to do a lot of mobile dictation, I'd recommend picking up Irradiated Software's drop-dead easy DropVox for $1.99 -- forget transferring your files by sync or by click, they'll just show up automatically in your Dropbox folder ready for transcription. (There's no step 3.) It remains to be seen how much of an impact the new on-device dictation capabilities will have on the pro-level dictation and transcription software market, but if you're already a MacSpeech Scribe user then it's worth giving Recorder a try... that is, if you're not already feeling silly talking to Siri.
Snail Mail My Email outsources your emotions to foreign hands
We bet the Britney Spears' classic Email My Heart would take great offense (and potential intellectual property beef) to Ivan Cash's startup, Snail Mail My Email. The 25-year old entrepreneur and lover of the quaint, soon-to-be anachronistic form of communication quit his advertising day job in favor of an out-of-pocket, handwritten transcription service. That's right, Cash and his global network of volunteers painstakingly re-create your digital salvos with the flourish of awkward and potentially illegible penmanship for free. Before you rush to overwhelm his servers with epic, misspelled ravings, pay close attention to that 100 word limit -- do-gooders' hands get tired, ya know. It's a quirky approach to letting that special someone know you care, and a great way to say, "I hope while you're reading this you're no longer drooling or pooping in your pants." (Their words, we swear!)
YouTube brings human-enabled closed captioning to live video for Google I/O
If you were glued to your computer during the live broadcast of the Google I/O keynote yesterday morning, you might have noticed a new feature accompanying an otherwise recognizable YouTube video. The online video provider used this morning's conference kickoff as the springboard for its live captioning feature, which brings human input to the transcription process. According to Google's Naomi Black, a team of stenographers banged out translations during this morning's keynote. The resulting captions were then displayed on the conference floor and delivered by an "open source gadget" to the I/O YouTube channel. This new feature apparently prevents the inaccuracies experienced using Google's automatic captioning function, which, if you'll recall, provided us with at least a couple hearty chuckles when we took it for a spin. The code behind the new live captions will be available to YouTube's partners and competitors on Google Code. You can check out tomorrow's keynote to see how the humans fare.
T-Mobile leak divulges return of unlimited WiFi calls, may add Name ID and Voicemail-to-Text
What's shaping up to be an epic week in tech news may be about to become even more exciting for T-Mobile fans. Internal employee docs are giving out some serious vibes that the company is ready to push out three important features to many of its phones as early as tomorrow. The first one to put a smile on your face is unlimited WiFi calling, which should be available as a free add-on to the Even More, Even More Plus, and 4G Do More plans. We're glad to see the service come back as a freebie, much better than the $9.99 per month asking price when it was hotspot@home. As if that isn't good enough by itself, the other services getting prepped for tomorrow's lineup include Name ID -- a caller ID service that shows the name, number, city, and state of anyone not listed in your contacts -- and Voicemail-to-Text, a new enhancement to the existing Visual Voicemail service that transcribes the full message into text form on select devices. Keep in mind that while these docs certainly do look official, it's all mere speculation until we hear actual word from T-Mobile about these new programs. With that said, we've got screenshots above and below, so feel free to glean as many details as possible from them.
Nuance opens Dragon Mobile SDK to app developers, we see end to embarrassing dictation
There are some messages that are just too embarrassing to dictate to a human being. Lucky for us and the retired circus contortionist we hired to type up our missives, Nuance is expanding the reach of its transcription software by making its Dragon Mobile SDK available to developers for use in iOS and Android applications. The SDK, which is free to members of the Nuance Mobile Developer Program, sports speech-to-text capabilities in eight languages and text-to-speech in 35. There are already apps out there that can do the job, including Nuance's own Dragon Dictation, but we welcome new advances in automated transcription. You know, it's not exactly a walk in the park dictating an entire Clay Aiken Fan Club newsletter to a guy named Sid the Human Pretzel.
OnStar announces Bluetooth voice app, reads your Facebook messages to you
Texting while driving is deadly, for serious. But, letting someone else read to you is rather less risky, and talking isn't so bad either -- in moderation, anyway. Bring those two together and you have OnStar's solution, an upcoming Bluetooth app that will read text messages and status updates to you and, somewhat more interestingly, lets you speak a custom message that will be transcribed to your recipient. Fascinating? Absolutely, but we can't wait to hear what sort of fun and cheeky mistranslations come out of that feature. You can also post voice messages to Facebook and say things like "call back" to return a call. The app is, as of now, intended for Android devices only and is set to hit the Market sometime in the first half of the year, and at least initially it'll only work on cars that have Bluetooth or those equipped with the company's new aftermarket mirror, though you'll have to be paid up on your OnStar dues if you want to use it. Full details in the PR after the break. %Gallery-112593%
AT&T launching voicemail-to-text service, new Mobile TV stations, Canada plans next week
This coming Sunday marks a straight-up bonanza of new services and tweaks from AT&T -- and while it may not combat a heavily-armed invasion of sentient handsets running Android, it's a nice little win nonetheless. Here's what we've got on tap: Voicemail to Text: This is a variation on a theme that has launched countless times both on other carriers and in the aftermarket, but AT&T's version is explicitly stated "not to be a replacement for a transcription service" because each message is limited to 60 seconds. Users have the option of routing messages to SMS, email, or both for a charge of $9.99 a month. Unfortunately, moving from basic voicemail to this new service will cause all existing messages to be lost, so be careful when adding this one to your plan. AT&T Nation with Canada: It's exactly what it sounds like -- AT&T Nation plans with a little extra Great White North thrown in for good measure. No long distance charges on calls to Canada, 1,000 night and weekend minutes that work in both countries plus full rollover and anytime minute compatibility; A-List and early nights / weekends can be added as well. New Mobile TV channels, coverage, and pricing: Three new channels will be added into the MediaFLO-based Mobile TV mix, though AT&T's being coy about what they are; all we know so far is that there's a comedy station, a "national broadcaster," and a kids' channel. Three new markets are launching between now and December 11, and seven more have launched since September 25. The biggest news here, though, might be that service is dropping from $15 to $9.99 a month, while Mobile TV plus unlimited data goes from $30 down to $24.99. It's still pricey, but it's an improvement. So, who's signing up for tiny teevee now that it's just a little bit cheaper? [Thanks, anonymous tipster] %Gallery-77322%
Friday Favorite: Transcriva
If you have a photographic memory, you may recall an article I wrote for TUAW about a year ago describing how to use AppleScript to make it easier to transcribe QuickTime movies and audio. In the comments for that piece, a program was pointed out to me (thanks imnotjesus) which has become a valuable tool in my toolbox. Transcriva is a single-purpose program for transcribing video and audio clips with a rich set of features certain to make your life easier. If you're doing professional transcription, recording audio notes in a class or a meeting for later reference, preparing sub-titles for a movie, or anything which involves copying what's being said or shown into text form, Transcriva has tools to fit, and pricing I find very reasonable. The main window of Transcriva offers a library view of your transcriptions, a media playback bar and your current transcription. With user-configurable keyboard shortcuts, it's possible to comfortably operate during a transcription without your hands ever needing to leave the keyboard. It even works with a foot pedal, if you're set up with one. You can control playback speed and set it to match your typing speed, as well automatically jump back a configurable number of seconds when you pause and resume playback. Of all of the features available, Follow-Along is my favorite. It allows you to play back your audio after you've transcribed it, and highlights the appropriate sections of the transcription as the playback head moves through them. More importantly, clicking on an area of the transcription jumps to its related point in the playback, allowing you to quickly review the audio associated with a note or transcription. This is important because that's exactly how I use Transcriva, taking notes from audio recordings or even during a recording when I'm using the built-in record features. Then I can review my hastily typed notes and immediately hear the audio that was happening at the time I took the note. It's great for recording meetings and annotating recorded Skype conversations. I imagine it would be an amazing tool in class, if you were in a situation where recording and typing were allowed. I haven't been to school for a while. The functionality is similar to Pear Note, but at $29.99US, Transcriva comes in $10US cheaper and packs more features. Transcriva can handle just about any type of audio or video you can play on your Mac. It uses QuickTime, and with Flip4Mac and Perian installed, you can extend the possibilities to include WMV, AVI, DIVX, FLV and more. When you're done with a transcription, you can export it to RTF or Word formats for sharing, publishing or continuing editing externally. I use Transcriva to recap interviews I do over Skype, and take my notes in an "outline" format which I can, with a little finagling, turn into a mind map or outline for an article. Transcriva has made my life exponentially easier and is a tool I'd gladly recommend to anyone with similar needs. My direct experience with the developer has also been great, with quick response times and a single bug report resulting in a new build within a couple of days. Transcriva is free to try, $29.99US to buy. You can download the trial at the Bartas Technologies site. If you hurry, it's even discounted to $19.99US in the MacUpdate Promo today.