An instant-translation device earpiece or headset has long been the stuff of sci-fi speculation. But for tech startup Waverly Labs, the future is now.
At this year’s CES, Waverly Labs introduced a suite of new services and hardware designed to provide near-instantaneous communication among multiple languages and dialects. Among them are Subtitles, a two-sided counter display for in-person interactions; Audience, a service designed for speakers and lecturers presenting before an auditorium; and Ambassador Interpreter, a recently-updated over-the-ear device that translates audio and text almost immediately.
“We are on a mission to bridge global communications by building a world without language barriers,” said Andrew Ochoa, Founder and CEO of Waverly Labs, in a press release. “We built a core translation technology for seamless conversations and are delighted to showcase Subtitles and Audience, our latest solutions, at CES along with the enhanced Ambassador Interpreter. While leveraging the same tech architecture, each solution’s hardware and experience are tailored to its unique use case.”
Subtitles is designed to provide seamless transactions and service exchanges between different language groups. Designed as a two-sided 3840 x 1100 screen, the device’s mic array picks up conversation between two parties, sends it to the cloud for translation, and flashes the processed text to the screen. It’s almost like watching a subtitled movie, but at the bank, hotel, airline, grocery store or any other organization or business.
A similar technical process is used for Audience, which allows lecture and presentation attendees to receive their own translations on their mobile device. Speech from the speaker’s mic is translated in the cloud and sent to each listener in their chosen language, accessed through a unique code or QR code. They can view the translation as text or hear it as audio.
It all comes bundled with support for over 20 languages and 42 dialects, including English, French, German, Italian, Portuguese, Spanish, Arabic, Greek, Russian, Hindi, Turkish, Polish, Chinese Mandarin, Japanese, Korean, Cantonese, Hebrew, Thai, Vietnamese, and Dutch.
Finally, the newly designed Ambassador Interpreter features a slimmer build and enhanced speech capture. It allows the wearer to navigate a multilingual environment, mastering each interaction with quick, machine-translated audio. It features three modes: a listen mode that picks up speech in a selected language up to eight feet away, a lecture mode for group settings and a converse mode for interactions with other Ambassador Interpreter users. The Ambassador Interpreter is priced $179.
According to Sergio Del Rio, Waverly Labs Founder, it was a mad dash to fine-tune the products in advance of their CES showcase several weeks ago.
“It took us about four months,” he said. “We did several sprints to get everything ready for CES. We have a small but talented team, and we managed to pull it off relatively quickly. We wanted to keep updating our current product Ambassador, so we had to do Subtitles and Audience as side projects. There was a remote conversations solution we had to table due to the lack of time though, but in retrospect, it was a wise decision.”
In many ways, the Waverly Labs team feels they have a product whose time has come. It took a long time for internet and computing technology to catch up with the vision for a near-instant translation device.
“I think we have enough hours of recording for training the Speech Recognition Models, the latency in the network is really good, the compression models have almost no loss in content, also the world has been slowly embracing a ‘Not only English’ approach,” said Del Rio.
Given that translation technology has been a staple of science fiction for ages, it’s no surprise that the Waverly Labs had plenty of brainstorming material when turning writers’ imaginations into reality. The team drew inspiration from a multitude of sources.
“The first industrial designs for Ambassador were mimicking fish, as we do have a lot of inspiration from Hitchhiker’s Guide to the Galaxy, but also there are many modern sci-fi productions that use real-time translation like Space Sweepers from Netflix,” Del Rio said. “It’s cool to see the normalization of this technology in popular culture.”
Of course, the sky’s the limit when it comes to expanding what wearable translation technology can accomplish.
“We are excited about simultaneous translation, meaning getting as close as possible as to real-time,” said Del Rio. “There is a lot of progress in custom terms and updated expressions and vocabulary, the new adaptation of TTS voices is exciting too, especially for games and the Metaverse. There are many new exciting components and [micro-electromechanical systems] that can enable us to improve the translation experience a lot while requiring less energy and performing better.”
That said, Del Rio said that multilingual humans won’t be obsolete any time soon. A human translator is the best bet for ensuring a seamless, nuanced communication experience — and it just so happens Waverly Labs’ products have the potential to be excellent language-learning tools.
“It will be very easy to get exposed, practice, and learn new language skills with these tools. No matter how fast we can do the translation, nothing will be faster than the human mind (at least for now),” he said.