texttospeech
Latest
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.
Google acquires speech synthesis outfit Phonetic Arts, plans to use Jack Donaghy's voice for everything
Don't ever knock Google for not reinvesting a little of that cheddar it's stacking in Mountain View. Barely two months after pulling the trigger on BlindType, El Goog has now sunk an undisclosed amount of money into Phonetic Arts, described as a speech synthesis company based in Cambridge, England. Naturally, Google's been toiling around the clock in an effort to better its speech technologies, and it looks as if it could be cutting out quite a few months (or years) of work with this one purchase. Phonetic Arts was known for being on the "cutting edge of speech synthesis, delivering technology that generates natural computer speech from small samples of recorded voice," and we get the impression that the team will be given clearance badges to enter Google's London-based engineering facility shortly. The company's own Mike Cohen is hoping that this will help us "move a little faster towards that Star Trek future" -- frankly, we're hoping to have Jack Donaghy's voice become the de facto standard in under a year. We hear some dudes at 30 Rock are already toying with a prototype...
On the road with the latest Navigon MobileNavigator
The iOS 4 update to MobileNavigator North America has been eagerly awaited by fans of the best selling turn-by-turn navigation system, and overall I'd call the update a win. To test the system, I took a drive from Southern Arizona to Chandler, a suburb of Phoenix. First off, having the multitasking capabilities was wonderful. I could take or make calls without stopping the route guidance, and while I was on the phone, I could get notice of turns. Since my phone was connected to my car via Bluetooth, the voice announcements came through the car speaker. I set up speed limit warnings, and the system dutifully warned me if I was going too fast, which is easy to do on those long, straight Arizona highways. %Gallery-97295%
Motorola announces H17txt Bluetooth headset with MotoSpeak
Looking for another reason to wear a Bluetooth headset all the time? Then you might want to consider admitting you have a problem, or you might want to get Motorola's new H17txt model, which can be paired with the company's new MotoSpeak text-to-speech application to read text messages aloud as they come in. That software is available for Android and BlackBerry initially, and will apparently even translate 150 different commonly used acronyms so things like l8r don't get read as l, eight, r. As for the headset itself, it's a fairly standard affair, with it boasting CrystalTalk noise-canceling technology, five hours of talk time and seven hours standby, and Bluetooth 2.1 compatibility. Still no indication of a price, but it should be available through Verizon starting today.
$2 Sensory chip could give toys (and other products) improved speech recognition, additional capabilities
Sensory Inc. may stay behind the scenes most of the time, but the company's speech recognition chips are already used in toys from JVC, Mattel, Hasbro and others, and it's now announced a new chip that could lead to toys with some significantly improved capabilities. Costing just $2 apiece (in quantities over 100K/year), the company's NLP-5X chip not only boasts support for speech recognition and text-to-speech that lets it "generate thousands of voices on the fly," but support for sound samples and MIDI playback as well. What's more, the chip uses what's described as an "incredible algorithm" that allows it to be on all the time and simply listen and activate itself when needed -- or when you least suspect it. Of course, while toys are one application, the company also sees the chip being used in a whole range of other consumer electronics -- Sensory even gives the example of an internet-connected oven that could let look up a recipe and then have a conversation with your oven about how you'd like to cook it.
ASUS DR-950 9-inch touchscreen e-reader brings text-to-speech and internet browser
ASUS is suddenly all chatty with its plans to enter the e-reader market in 2010. Just yesterday we got word of a 6-inch color (claimed to be OLED by InGear) e-reader from ASUS by the name of DR-570 headed to retail before the year is through. Now we've got details of a second ASUS e-reader, dubbed the DR-950 that should arrive sooner. This time we're looking at a 9-inch Sipix panel with 1,024 x 768 pixel resolution pushing 16-levels of gray just like the Jinke reader unveiled at CES. The touchscreen DR-950 features text-to-speech (based on Svox engine supporting 26 languages), a web browser that works in portrait or landscape modes, a virtual keyboard and handwriting input, a RSS reader, and dictionary (with expandable database) with real-time translation. Spec-wise, the 222 x 161 x 9-mm / 370-gram reader packs WiFi and HSPA (WiMax is optional) data radios, 3.5-mm headphone jack and stereo speakers, with 4GB of internal memory and SD Card expansion. Supported formats include PDF, TXT, Audible, MP3, and unprotected ePub. Not bad ASUS, not bad. Now let's see some content partners, eh? See it pictured browsing the web after the break.
Teclast enters the e-book fold with the K3 Talking Portable Library
Judging by the snowballing number of e-readers we've seen over the last year or so (culminating at CES 2010, as you are no doubt well aware) either people really, really, really love these things, or OEMs are going for the lowest hanging fruit. And what do we have here from Teclast? The company, known 'round these parts for its portable media players, has entered the game with the K3. In addition from the usual array of supported document formats this bad boy rocks a 6-inch E Ink display, an MP3 / MP4 player and -- hold on to your caps! -- text-to-speech either in English or Chinese. The term the company is using to market this bad boy is "Talking Portable Library," but since most TTS devices usually make HAL sound like James Earl Jones we're going to suggest that anyone truly interested in this technology stick with the Kindle. Besides, Tom misses you! You two haven't talked in a long time.
Amazon beefing up Kindle's functionality for vision-impared users as B&N's Nook stays silent
While the Amazon Kindle's text-to-speech functionality might seem like a gimmick for some, it's anything but for blind, vision-impaired and dyslexic users. Unfortunately, the device's accessibility so far hasn't extended to the menus. That's set to change next year, however, with Amazon promising to release an audible menuing system for navigating the unit look-free. Amazon's also prepping a new "super size" font, that doubles the current largest font in height and width. It all sounds great, but it also seems like a subtle dig at Barnes & Noble, whose brand new Nook reader is skipping out on text-to-speech (for this generation, anyway). Barnes & Noble claims that it's due to the sub-par experience on "other devices," but for now that means the Kindle might just be most accessible dedicated e-reader around -- at least once this new software rolls out, supposedly by summer 2010.
Intel Reader handheld device converts text to spoken word
While we busy ourselves with arguments about how to replace the perfectly usable book gadget, Intel is dealing with the complications that keep far too many people from being able to read at all. Presenting its Reader as a necessity rather than luxury, Intel has shown off its vision for how visually impaired and dyslexic people can obtain access to the written word. Combining a text-scanning camera with a text-to-speech engine (powered by an Atom inside) is certainly no bad idea, but as the video beyond the break will show you, Intel's execution isn't exactly stellar. The arrhythmic, robotic reading from Alice In Wonderland left us shaken, but what floored us was the $1,499 asking price. Yes, it's a customized niche device, but we reckon we could build something similar for a third of the price. Full PR and video after the break.
Random House now disabling text-to-speech function of Kindle e-books
The much-touted and extremely controversial story of the text-to-speech function of Amazon's Kindle 2 could fill a very large e-book. The tale continues to get longer still, as at least one major publisher -- Random House -- has thrown the dreaded "kill switch" on about 40 of its titles, including authors such as Toni Morrison, and, ironically, Stephen King (who you will remember was part of the Kindle 2's launch). Random House disabled the function without much fanfare, or an official announcement, but you can be sure this isn't the final chapter.
Mac Automation: Make your text speak its mind
Have you ever wanted to type something into your Mac and have it record what you typed audibly using any one of the Mac OS X voices? It's easily done with a few Automator actions, and in this Mac Automation post, I'll show you how. Open Automator and find the following actions, then drag them to the workflow area (in the same order): Get Specified Text Text to Audio File Running the workflowWith these two actions in place, you can enter text into the Get Specified Text action, and a file name in the Text to Audio File action, and run it with great results. The text that you enter will be magically encoded into audio using the specified voice in the audio file action. Read on to learn more about saving the workflow, and adding additional options.
Is this the voice behind Kindle's text to speech feature?
The controversy surrounding the Kindle 2's text to speech functionality continues unabated, and as we delve further into the legal back 'n forth between Amazon and the Authors Guild, we wonder if we haven't discovered the true identity of "Tom." According to David Pogue, the New York Times journalist has received an email from a certain adult contemporary heartthrob named Tom Glynn, claiming to be the source of the mysterious voice. What do you guys think -- would you pay $359 for this guy to read you a bed time story?
Authors Guild president: Kindle's a swindle
We're not sure his tone or rhetoric are the most convincing, but Authors Guild president Roy Blount Jr. has a little op-ed in the New York Times today attempting to explain his organization's objection to the Kindle 2's text-to-speech features. If you read our little Know Your Rights piece, you already know what he's on about: the main problem is that Amazon isn't paying for both ebook and audiobook rights for Kindle content, and Roy's worried that eventually computerized text-to-speech will be good enough for consumers to eschew buying audiobooks entirely. Take a deep breath, count to ten, consider that audiobooks are a billion-dollar business, and you can sort of see where the Authors Guild is coming from -- Roy doesn't sound too crazy when he says he thinks "authors have a right to a fair share of the value that audio adds to Kindle 2's version of books."What does this mean for you? Well, probably just higher prices in the short term, as we're guessing publishers will start increasing ebook license fees to cover what they think they're going to lose on audiobook sales, and Amazon and other ebook retailers will just pass those costs along. Lame, sure, but it's not the crackdown some were hysterically predicting -- Roy's pretty clear that the Authors Guild doesn't care about parents reading to kids or text-to-speech for the blind, just the Kindle's impact on the audiobook market. We'll see how Amazon and the Guild resolve this one over the next few months -- in the meantime, point your Kindle to the read link and blow Roy's mind by having Tom read the op-ed to you.
Kindle's text-to-speech feature voiced by "Tom"?
The controversial text to speech feature of the Kindle 2 has also been one of its most touted. Well, a dedicated investigator over on the KindleBoards has unearthed the purported true identity of that buttery smooth Kindle 2 voice: and he's called "Tom." Tom seems to have been developed by Nuance Technology, and if you hit the read link, you can hear comparisons of the actual Tom, and the Kindle 2 voice, then form your own opinion -- have we found him out, or is he still shrouded behind a thick curtain of mystery?
Know Your Rights: Does the Kindle 2's text-to-speech infringe authors' copyrights?
Know Your Rights is Engadget's technology law series, written by our own totally punk ex-copyright attorney Nilay Patel. In it we'll try to answer some fundamental tech-law questions to help you stay out of trouble in this brave new world. Disclaimer: this isn't legal advice, but it is best read aloud by a text to speech app. Hey, so does the Kindle 2's Read to Me text-to-speech feature really infringe on authors' copyrights? It's nice to be back! It's been a while. Yeah yeah. Get to it. Okay, so the issue is that the Kindle 2's Read to Me feature obviously threatens the audiobook market, and while at first blush it seems like the Authors Guild has a pretty weak case when executive director Paul Aiken says things like "They don't have the right to read a book out loud," it's not necessarily as ridiculous as it seems.
Mac 101: Make any text speak to you
Have you ever wished your Mac could read a long text document to you? Well, with the speech service, you can easily have your Mac read as much or as little text as you want. While in Safari, TextEdit, Pages, and other applications; select the text you want to be read, then click the application name in the menu bar. Go to Services > Speech > Start Speaking Text. Your Mac will then use the default voice to read the text. It will continue reading until it reaches the end of the selected text, or you can select Services > Speech > Stop Speaking to end it immediately. You can change the default voice by opening System Preferences (Apple menu > System Preferences) and going to Speech > Text to Speech. Once there, select a voice from the drop-down "System Voice" menu.Want more tips and tricks like this? Visit TUAW's Mac 101 section!
GhostReader 1.5: text to audiobook
GhostReader is a text to speech application that allows you to have your computer read aloud PDF, Word, and text files and save the output in a variety of formats for importing into iTunes. While Mac OS X does have built-in text to speech capabilities in the Universal Access Preference Pane (VoiceOver), GhostReader adds more voices (including various voices tailored to multiple European languages), allows you to modify how words are pronounced, and automates the conversion to audiobooks.GhostReader 1.5 is $39.95 for one language; additional languages are extra; a demo is available.
Nifty hack adds text-to-speech to Garmin c510 / c530
Look, we fully understand that you're probably not ready to drop fresh coinage to upgrade your one-year young c510 or c530 navigation device, but why not treat it to a svelte hack that will have it acting like the new kids on the block? Yep, your aging Garmin can now sport the same text-to-speech capabilities as the newer c550 / c580 devices with just a little a good bit of internal tweaking. Considering that the elder machines boast less storage, yet TTS requires quite a bundle, you'll be forced to purge your GPS of unnecessary files in order to make it all fit, and aside from exposing yourself to the awful possibility of botching your portable tour guide, you'll also need to be comfortable with a long-winded process of implementing (and locking in) the changes. All in all, we can't complain with teaching old NAVs new tricks, but this one certainly isn't for the faint of heart. Hit the read link if you dare.[Via NaviGadget]
