This Startup Wants Translate Your Own Voice into Another Language that You Can’t Speak

BabelOn, a startup in the San Fransisco area, is developing software that can transform your speech from English to any other language, without using any additional translation services, and it will sound like you own voice. While using artificial means to create the sounds of a human voice, a technique called speech synthesis, has been around for a while, BabelOn is offering a very specific and unique spin on the technology. Using a specialized combination of custom designed hardware and software, BabelOn will analyze your voice for its unique characteristics then use those results to recreate language that sounds like the words are coming out of your own mouth in any language you want.

Originally the idea was conceived for use in film dubbing or translating video games but the ultimate goal for BabelOn is to provide real-time language translation in your voice, like when you’re on a Skype call or similar circumstances. Although Microsoft currently offers a comparable service for some time, their voice is digital sounding, like Siri, making the BabelOn difference more personal.

While this is certainly a very interesting concept, it is still very early on in BabelOn’s development. There has yet to be a software demonstration, nor have they done any work for clients. Currently, BabelOn is bidding for a soon to be released video game translation but it is not a done deal. The software has a potential for success but also presents a glaring security concern in the concept of having one’s voice “stolen”.

Although the idea behind this novel business concept came from co-founder Daisy Hamilton’s parents and they go a patent for the core tech back in 2004, it wasn’t until recently that the additional technology needed to make BabelOn function as fully imagined existed.

Now that the appropriately sophisticated hardware and software are available, makescan launch full steam ahead. It all starts by creating the BLIP, or BabelOn Language Information Profile. In a two-hour recording studio session, having read aloud specified scripts in various emotional states, a personalized BLIP is formulated

Hamilton sees the voice as more than a simple sound but more like a unique instrument. The fine-tuned hardware captures your breathing cadence, the sound of your voice from your chest and throat, the way your mouth moves, and a number other defining characteristics make your voice uniquely you. “It’s both visual and vocal feedback that’s captured into a single continuous stream,” Hamilton explains.

After your BLIP is created, BabelOn can take it, transform it into another language, and imitate a chosen emotion as required by script without needing a whole new recording session. Such a technology would be a large benefit for a film production or video game developer that needs to change the language of the project but don’t have a full budget for another set of native speaking actors or any more time for a recording studio.

In its early offerings, BabelOn will have a complement of languages including English, French, Spanish, German, Portuguese, Mandarin, Japanese, and Hindi, adding other languages along the way. However, there are limitations to what BabelOn will be able to do. For example, it cannot do both translation and voice creation. The program must be given either an already written script content in the desired language but you can choose any emotion for the new recording to have. Hamilton calls it “emotional markup language.”

For the hardware, BabelOn partnered with Lawrence Livermore National Laboratory, a federal institution that works on developing science and technology in the U.S. The hardware is a type that was already being used by the U.S. Department of Defense in other applications.

According to Hamilton, it can take hours to completely process a full script into a different language but as the software and computer power improves, she hopes to have the system running in almost real time. That capability will expand BabelOn’s market base beyond dubbing work and make it practical for more users to have a personal and emotive multilingual expression.

The real challenge is the security concerns previously mentioned. If the BLIP technology becomes available to many users and applications, couldn’t that allow hackers access to manipulate anyone’s voice? While the company maintains an ethics board to oversee company use of the technology, but that doesn’t prevent an outsider perpetrating digital theft.

Hamilton has claimed that BabelOn will ”use a highly encrypted offline voice vault to store all of the BLIP, which would be curated upon request of the [original] speaker.” Storage offline does make the voice archive more difficult to access via cyber intrusion and Hamilton added that if any thing is altered in a voice, there are visible markers. Although no on can be sure on how the service will operate at larger scales if it becomes popular, BabelOn is already conscious of future customer security concerns.”Security of BLIPs is massively important to us, as we’d never want to threaten someone’s vocal authenticity,” she said.

Despite their best efforts, however, security may still be BabelOn’s biggest stumbling block to wider commercial use.  When used for script recording, whether movies or video games, a BLIP can be deleted or destroyed immediately after it’s served its purpose. But a real world test of this powerful tool has yet to be done and the ability to control the voice of another could be a substantial lure for hackers.

Currently, BabelOn is seeking to fund by crowdfunding via Indiegogo, even though it won’t be available to the mass public market yet. However, Hamilton says the campaign is to fund the license on the software while more work is completed and there is a backup plan if the funding campaign goals are not met. She adds that “It’s just as much about using Indiegogo as a launch pad to put BabelOn out in the world.”

Overall, Hamilton remains confident about the future of BabelOn and having their first client soon. Once the video game is translated using BabelOn technology, it can be the company ambassador, showing the world what it can do. Until then, it’s a compelling idea and who knows how far it can go in the film industry, with video games, and even as part of consumer based apps in the future.

More News to Read