Speech Used to Pimp News

Tech startup PimpMyNews.com announced its wide launch today. Dubbed the world’s first "talking social news site," PimpMyNews speechifies newsfeeds and blog entries from around the Internet using text-to-speech technology from a variety of vendors, including NeoSpeech.

After signing up, users customize their topics from 57 categories, including entertainment, politics, and business. Users with iPods or other MP3 players can elect to receive fresh content regardless of location—allowing them to stay up-to-date even without a computer. For PimpMyNews co-founder John Atkinson, audio gives users the freedom to consume media whenever and wherever they want. 

"To me, voice adds a portable, easier consumption method than reading," he says. "The Internet has grown in popularity." He cites a Pew Research Center study which states that one-in-three Americans get their news online today, versus one-in-fifty a decade ago.  "The problem with using the Internet to read news is that in most cases," he continues, "you’re searching for news on multiple sites and have to sit and be at your computer. The nice thing about PimpMyNews is we convert that news to a portable audio format. It’s sort of like TiVo for news and blogs."

It’s still too early to tell how PimpMyNews will be received by the general populace.  While it certainly provides constant connectivity—it updates frequently from a variety of sites, including newsfeeds like Reuters and blogs like Gizmodo—it’s unclear how users will respond to the synthesized voice. For this writer, the voice was clear and distinct when it read hard news stories, but didn’t fare as well reading content from blogs like PerezHilton.com, where the voice was unable to capture the ironic underpinnings of the site.     

"Text to speech has for a long time been monotone," says Atkinson. "It’s very difficult to automate that on a regular basis. And everything you see on PimpMyNews is fully automated, happens on the fly in real time. We’re pretty pleased with our ability to get content and convert it in real time, but user’s ears are very finely tuned." 

Still, the site is in its early days and, besides the synthesized voice, Atkinson and his crew continue to refine the site. For instance, to avoid mispronunciations—inevitable in a world where Mahmoud Ahmadinejad is a prime newsmaker—PimpMyNews maintains a customized dictionary that’s constantly updated using either visual or phonetic representations. Additionally, users can write in to identify mispronunciations.

So far, Atkinson describes the site as a "labor of love" though the plan is to get revenue through advertising and premium services that users have to pay for—such as an intelligent search that monitors a user’s specific area of interest.  

Voice has yet to find its widespread function on the Internet, which thrives on mostly graphical interfaces. Recently however, sites like Facebook and Myspace have initiated applications that give voice a greater role. PimpMyNews also gives users the ability to consume media in a unique way.  And while analysts predict that voice will soon be integrated more fluidly into the Web, they’re unsure how and in what form that integration will take place.   

SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues