
Linguists estimate that at least half of the world's estimated 7,000 spoken languages will become extinct by the century's end, due to forces ranging from globalization to cultural assimilation.
Part of the challenge of documenting and revitalizing endangered languages is a lack of texts and speech recordings to work with. Seneca, a language of one of the six Iroquois Nations in North America, has only about 100 first-language speakers and several hundred more second-language learners.
Automatic speech recognition (ASR) technology is widely used to transcribe languages with millions or billions of speakers, like English and Mandarin. But it has only scratched the surface with languages like Seneca, which have vastly fewer speakers and significantly less data to work with.
Now a team of researchers at the Rochester Institute of Technology in New York, along with colleagues from the University at Buffalo, is tapping deep learning to bolster the ability of ASR. And while its focus is on Seneca, the project's vision encompasses the preservation of languages globally as well as an important part of our shared cultural history.
Knowing about different languages teaches us a lot about how our brain works, said Emily Prud'hommeaux, an assistant professor of computer science at Boston College and a research faculty member at RIT. When you document a language, you're preserving information not only about that language but also about how humans use language in general.
It's no coincidence that Prud'hommeaux and her team started with the Seneca language. Three members of the Seneca nation are part of the effort - a direct connection that is rare in research of this type, she said.
Leading the charge is Robbie Jimerson, a Ph.D. student in RIT's Golisano College of Computing and Information Science. He is a member of the Seneca Nation of Indians and is passionate about ensuring the survival of the Seneca language.
There's a big effort by the leaders of the tribe to preserve and promote our language, said Jimerson. I was looking for an opportunity to contribute.
Using GANs to Create More Language Samples Now in its third year, the project has had challenges when it comes to accumulating language data. Jimerson said the Seneca community can be guarded about what it shares with other people, so there wasn't an abundance of recordings of the language being spoken. He set out to change that.
He started by recording friends and elders who speak the language and asking them to record their friends. He found out whenever someone was speaking Seneca in public. He asked for family recordings of elders telling stories handed down from previous generations. And he grabbed any publicly available videos or recordings he could find online.
The team has fine-tuned an ASR model for Seneca, running it through generative adversarial networks to create more samples out of the limited number of recordings. The model turns wave files of the spoken language into streams of characters, while computing probability and making corrections.
The resulting data is fed into a deep learning model that in turn expands upon the ASR model's accuracy.
The team's networks run in two compute settings: on a nine-server machine learning lab running a variety of NVIDIA Tesla GPUs, and on a university cluster of large servers, each running 10 NVIDIA Tesla P4 GPUs. Each cluster runs a range of deep learning frameworks such as TensorFlow and Caffe.
The computer engineering cluster is for all students in the computer engineering department, and so they have to compete' for these resources, said Ray Ptucha, assistant professor of computer engineering at RIT, another collaborator on this project.
With access to these clusters at a premium, Jimerson tests code and checks the stability of models on a local machine running an NVIDIA TITAN X rather than inconvenience other students by running a model that might crash.
Achieving Better Accuracy So far, the team's efforts have brought the word error rate of its ASR model from 70 percent down to 56 percent. The goal, said Prud'hommeaux, is to get that rate down to 25 percent, which is where ASR systems were in processing English several years ago.
The more samples of spoken and written Seneca the team can accumulate, the more the error rate will decrease. (Today, English ASR models can achieve word error rates as low as 5 percent.)
The team's work is expected to help with language preservation efforts around the world.
Prud'hommeaux said the team has an agreement with an archiving institution that's a condition of a grant the project received from the National Science Foundation. The resulting language archiving database will be made available as a resource for other efforts seeking to document threatened languages.
Additionally, Prud'hommeaux said the team's work could prove helpful for any deep learning effort that has to make do with limited amounts of data.
Read more about the team's work in their research papers here and here.
Feature image: The Haudenosaunee (Iroquois Confederacy) flag, via Wikimedia Commons.
Most recent headlines
05/01/2027
Worlds first 802.15.4ab-UWB chip verified by Calterah and Rohde & Schwarz to be ...
07/10/2026
Dalet, a leading technology and service provider for media-rich organizations, today announced the latest Long-Term Supported (LTS) release of Dalet Flex. Build...
06/09/2026
June 9 2026, 23:00 (PDT) Dolby and MagentaTV Bring Fans Closer to the FIFA Worl...
28/08/2026
Encompass Digital Media and Grass Valley are expanding their collaboration with the launch of Altitude Playout Plus, a software-defined channel origination plat...
28/08/2026
Yospace, the trusted leader in Dynamic Ad Insertion (DAI), has announced the release of its Ad Presentation solution, which enables broadcasters to apply non-li...
28/08/2026
TSL Showcases Reduced Complexity for Control & Monitoring Across IP Workflows at IBC2026
New TSL Hummingbird capabilities including Dante control routing, toge...
28/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
28/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
28/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
28/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
28/08/2026
GlobalM and Lamhas Satellite Services today announced a new agreement that will see GlobalMs GMX technology deployed as part of Lamhas expanding managed IP broa...
28/08/2026
Integrated 25GbE Connectivity Delivers UHD Video Directly to Up to Two Monitors, Eliminates External Gateways, and Supports SMPTE ST 2022-7 Network-Path Protect...
28/08/2026
Interactive Ask Me Anything sessions will connect broadcasters and technology providers with the experts developing a practical, interoperable roadmap for sof...
28/08/2026
At IBC2026, TAG will show sub-2-frame latency and 800 Gbps of aggregate scale for uncompressed ST 2110 monitoring in software, live MXL interoperability with pa...
28/08/2026
SportsBox, BHV's Ultra-Compact 4K Remote Switcher, Moves from Development to Production at IBC 2026
SportsBox brings flexibility without compromise to high...
28/08/2026
Lutz Rathmann named CEO of Riedel Communications Anthony Zuyderhoff to lead unified customer-facing organization as CCO
Riedel Communications is strengthenin...
28/08/2026
LA JOLLA, CA-Scripps Research has established the Pete Schultz Fund for Innovati...
28/08/2026
LA JOLLA, CA-Chemical structures called cyclopropanols are among the most useful...
27/08/2026
ESPN and Disney are launching College Football Saturdays, adding two college football games per week to Disney for U.S. subscribers as part of their existing ...
27/08/2026
Zixi will exhibit at IBC2026 (Hall 5, Stand 5.A76, RAI Amsterdam, September 11-14), demonstrating the Zixi Platform and the latest release of Zixi Broadcaster, ...
27/08/2026
NDI has announced that its ecosystem has surpassed 1.2 million NDI-enabled hardware nodes worldwide, spanning products from more than 200 hardware brands across...
27/08/2026
Clear-Com conducted a pre-launch field trial of FreeSpeak Cell during Carnival 2026 in Rio de Janeiro, in partnership with Brazilian systems integrator Savana C...
27/08/2026
BHV returns to IBC 2026 with an important milestone for its own SportsBox: the broadcast industry's smallest remote vision mixer and format processor has mo...
27/08/2026
TOGETHXR and Snapchat have announced the second cohort of their Snap the Gap content creator program, welcoming nine athletes as Snap Stars. The program provide...
27/08/2026
MediaKind has announced that CNBC UK has selected Beam to modernize its contribution and distribution infrastructure for live broadcast operations. The deployme...
27/08/2026
Monitor engineer Chris Dwight has deployed Sennheiser's Spectera wideband bi...
27/08/2026
Telestream has announced an expansion of its UP cloud-native platform, adding capabilities for orchestrating Telestream services, third-party technologies, and ...
27/08/2026
Swerve TV has signed a multi-year partnership with GLORY Kickboxing, bringing live GLORY events and library programming to Swerve Combat, Swerve TV's free a...
27/08/2026
Dalet has announced a new integrated content experience with Agence France-Presse (AFP) in Dalet Flex, designed to bring AFP's editorial metadata into the p...
27/08/2026
Grass Valley will demonstrate an integrated live production environment at IBC2026 (Booth 9.A01, RAI Amsterdam, September 11-14), combining K-Frame, Maverik X, ...
27/08/2026
The National Women's Soccer League and Roku have announced an expanded partnership, with the Roku Sports Channel becoming the exclusive streaming home of ei...
27/08/2026
FuboTV has announced the launch of ACC Network Extra (ACCNX) and SEC Network (SECN ) on the Fubo streaming service at no additional cost to subscribers with En...
27/08/2026
Sports marketers have entered the experiential age. Today, sports teams, leagues, and venues are increasingly using immersive interactions and real world events...
27/08/2026
MultiDyne will introduce the MDoG-8800 at IBC2026 (Stand 11.C12, RAI Amsterdam, September 11-14), extending its openGear portfolio with a high-density fiber tra...
27/08/2026
The Joint Task Force on the Dynamic Media Facility (JT-DMF), an initiative of AMWA and the EBU, will host a series of informal Ask Me Anything sessions during...
27/08/2026
Homestream, a mobile app that automatically syncs a team's local radio broad...
27/08/2026
A packager for TNT Sports, Kwokman Productions is on hand for the unique event w...
27/08/2026
After more than 3,000 games, five members of YES Network's original staff re...
27/08/2026
Physical modelling orchestral instrument announced
Following their recent cryptic hints, Dreamtonics have officially revealed the latest addition to their s...
27/08/2026
M5 Ultra-equipped Studio most powerful Mac to date
Apple have just announced a design refresh which they say makes the Mac mini fiercer and more versatil...
27/08/2026
Sampler receivers major upgrades
Bitwig's latest DAW update brings some major changes to the Sampler device, kitting it out with new slicing options, tw...
27/08/2026
Designed specifically for hard-hitting modern bass
Designed specifically for bass-heavy electronic music, the latest plug-in from Arturia aims to offer an a...
27/08/2026
Advertisers gain granular CTV targeting and turnkey reporting across eligible in...
27/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
27/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
27/08/2026
Profuz Digital makes a mark by launching a powerful upgrade to its Profuz LAPIS digital media management platform this IBC in Amsterdam at Stand 8.A88. The late...
27/08/2026
DHD.audio has chosen IBC 2026 as the launch venue for what it defines as one of the most radical advances in the companys 30 year history.
Every broadcast trad...
27/08/2026
Direct connectivity between planning, production and graphics for creativity and streamlined workflows...
27/08/2026
New Photoshop innovations bring you more choice and control at every stage of yo...
27/08/2026
First Two Seasons of Vertical Series Vultures Shot on Blackmagic Design Cameras
Brie Clayton August 27, 2026
0 Comments
Director and DoP of Vultures ...