
AI built for speech is now decoding the language of earthquakes.
A team of researchers from the Earth and environmental sciences division at Los Alamos National Laboratory repurposed Meta's Wav2Vec-2.0, an AI model designed for speech recognition, to analyze seismic signals from Hawaii's 2018 K lauea volcano collapse.
Their findings, published in Nature Communications, suggest that faults emit distinct signals as they shift - patterns that AI can now track in real time. While this doesn't mean AI can predict earthquakes, the study marks an important step toward understanding how faults behave before a slip event.
Seismic records are acoustic measurements of waves passing through the solid Earth, said Christopher Johnson, one of the study's lead researchers. From a signal processing perspective, many similar techniques are applied for both audio and seismic waveform analysis.
The AI model was tested using data from the 2018 collapse of Hawaii's K lauea caldera, which triggered months of earthquakes and reshaped the volcanic landscape. The lava lake in Halema uma u during the 2020-2021 eruption (USGS/F. Trusdell) is a striking reminder of K lauea's ongoing activity. Big earthquakes don't just shake the ground - they upend economies. In the past five years, quakes in Japan, Turkey and California have caused tens of billions of dollars in damage and displaced millions of people.
That's where AI comes in. Led by Johnson, along with Kun Wang and Paul Johnson, the Los Alamos team tested whether speech-recognition AI could make sense of fault movements - deciphering the tremors like words in a sentence.
To test their approach, the team used data from the dramatic 2018 collapse of Hawaii's K lauea caldera, which triggered a series of earthquakes over three months.
The AI analyzed seismic waveforms and mapped them to real-time ground movement, revealing that faults might speak in patterns resembling human speech.
Speech recognition models like Wav2Vec-2.0 are well-suited for this task because they excel at identifying complex, time-series data patterns - whether involving human speech or the Earth's tremors.
The AI model outperformed traditional methods, such as gradient-boosted trees, which struggle with the unpredictable nature of seismic signals. Gradient-boosted trees build multiple decision trees in sequence, refining predictions by correcting previous errors at each step.
However, these models struggle with highly variable, continuous signals like seismic waveforms. In contrast, deep learning models like Wav2Vec-2.0 excel at identifying underlying patterns.
How AI Was Trained to Listen to the Earth Unlike previous machine learning models that required manually labeled training data, the researchers used a self-supervised learning approach to train Wav2Vec-2.0. The model was pretrained on continuous seismic waveforms and then fine-tuned using real-world data from K lauea's collapse sequence.
NVIDIA accelerated computing played a crucial role in processing vast amounts of seismic waveform data in parallel. High-performance NVIDIA GPUs accelerated training, enabling the AI to efficiently extract meaningful patterns from continuous seismic signals.
What's Still Missing: Can AI Predict Earthquakes? While the AI showed promise in tracking real-time fault shifts, it was less effective at forecasting future displacement. Attempts to train the model for near-future predictions - essentially, asking it to anticipate a slip event before it happens - yielded inconclusive results.
We need to expand the training data to include continuous data from other seismic networks that contain more variations in naturally occurring and anthropogenic signals, he explained.
A Step Toward Smarter Seismic Monitoring Despite the challenges in forecasting, the results mark an intriguing advancement in earthquake research. This study suggests that AI models designed for speech recognition may be uniquely suited to interpreting the intricate, shifting signals faults generate over time.
This research, as applied to tectonic fault systems, is still in its infancy, Johnson. The study is more analogous to data from laboratory experiments than large earthquake fault zones, which have much longer recurrence intervals. Extending these efforts to real-world forecasting will require further model development with physics-based constraints.
So, no, speech-based AI models aren't predicting earthquakes yet. But this research suggests they could one day - if scientists can teach it to listen more carefully.
Read the full paper, Automatic Speech Recognition Predicts Contemporaneous Earthquake Fault Displacement, to dive deeper into the science behind this groundbreaking research.
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...
25/08/2026
Open, Sustainable, Predictable: What Utah Scientific Is Bringing to IBC 2026 As broadcast leaders prepare to gather at the RAI Amsterdam for IBC 2026, the main ...
25/08/2026
NC State's Matthew Byrd breaks down how an IP-based infrastructure is delivering greater flexibility and scalability for Wolfpack productions...
25/08/2026
Deal expands the DTC reach in-market of Capitals, Wizards, Mystics games and oth...
25/08/2026
The two-year renovation period will create a stadium of the future for the NFL f...
25/08/2026
Behind The Mic provides a roundup of recent news regarding on-air talent, includ...
25/08/2026
Shell Energy Stadium in Houston has installed a new L-Acoustics A Series loudspeaker system, designed and installed by LD Systems, a Clair Global brand. The 20,...
25/08/2026
FloCollege is expanding its distribution and product offerings for the 2026-27 season, covering more than 20 NCAA partner conferences across Division I, II, and...
25/08/2026
TNA Wrestling and REVOLT have announced a content partnership bringing exclusive TNA Xplosion matches and content from TNA's library to REVOLT.
Our indust...
25/08/2026
Gracenote, Nielsen's content intelligence business, has released its Q3 2026 Data Hub analysis showing that FAST sports programming is growing at more than ...
25/08/2026
Amagi has appointed Thomas d'H rouville as Head of Sales for France, Benelux, and Nordics. Based in Paris, he will lead Amagi's growth strategy across t...
25/08/2026
IPC has appointed CCATech as its channel partner in Brazil, expanding the company's presence in Latin America. CCATech was founded in 2023 by Carlos Abrah o...
25/08/2026
FanConnect has announced the FC4, a multi-output media player designed to drive mid-sized video walls, multi-screen concession stands, and multi-display digital...
25/08/2026
JWX has launched JWX Content Hub, a platform for publisher video that combines content transformation, social distribution, audience engagement, and monetizatio...
25/08/2026
Techex has appointed Stuart Almond as Chief Revenue Officer. Almond brings more than 25 years of experience across technology, media, and telecommunications, ha...
25/08/2026
Grass Valley will demonstrate infrastructure connecting SDI, SMPTE ST 2110, MXL, and software-based processing at IBC2026 (Booth 9.A01, RAI Amsterdam, September...
25/08/2026
Franchises are taking greater ownership of production, distribution, and the fan...
25/08/2026
The first-year circuit for former professional athletes leans on LiveU, Starlink...
25/08/2026
Announcing the 2026 Sundance Institute Breakthrough | Focus Features Fellows
L...
25/08/2026
Auction Part 3 running until 9 September 2026
Soundgas have announced that the latest stage of their John Paul Jones Gear Auction is now live, and that bids...
25/08/2026
Over 60 award-winning artists, songwriters, producers, engineers
The MONO Music Conference (formerly Music Expo SF), San Francisco's premier event for m...
25/08/2026
All three versions receive firmware upgrade
Groove Synthesis have just launched a free firmware update that kits the 3rd Wave out with two major new feature...
25/08/2026
eds3_5_jq(document).ready(function($) { $(#eds_sliderM519).chameleonSlider_2_1({...
25/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
25/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
25/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
25/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
25/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
25/08/2026
Lingopal, the AI-powered live translation and localization specialist, is unveiling new platform capabilities at IBC2026 (11 14 September, RAI Amsterdam), helpi...
25/08/2026
Bminty is supporting TV5MONDE in the evolution of its editorial processes through a new AI-assisted press information generation solution. Combining the Louise ...
25/08/2026
Two Cultures. Millions Reached. One Berklee Connection. Meet Annie Dickinson and Zhengxi Zhang-the couple who turned their creative partnership into music car...
25/08/2026
Amagi, the media industry cloud platform for unified broadcast, streaming, and monetization, today announced that Law&Crime has migrated its Court TV channel, s...
25/08/2026
Broadcast Solutions, a leading systems integrator and provider of innovative solutions for the broadcast and media industry, is showcasing the breadth of its ca...
25/08/2026
Thatcham, UK 25 August 2026: Test & measurement innovator, Leader Electronics, has announced that leading broadcast rental company Minitech has purchased a Le...
25/08/2026
Appear's X20 Platform enables visually lossless video contribution with sub-frame latency and full path redundancy for TVTEL's centralised VAR operation...
25/08/2026
World-first demonstration combines 5G Broadcast and Media over QUIC for seamless, ultra-low-latency hybrid media delivery
Big Blue Marble will showcase new dev...
25/08/2026
MNC Software, a global leader in network management and operational support systems tailored to the broadcast and media industry, will showcase its complete sof...
25/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
25/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
25/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
25/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
25/08/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
25/08/2026
Company to Demonstrate Practical AI Applications That Add Value Today and Longer-Term Path Toward Agentic AI
At IBC2026 (Sept. 11 14, RAI Amsterdam, Stand #1.B...
25/08/2026
Serving South Florida with a unique blend of jazz, Latin jazz, salsa, reggae, blues, and culturally reflective programming, WDNA 88.9FM operates as both a tradi...
25/08/2026
GlobalM announced that it is integrating its GMX software-defined media platform with Oracle Cloud Infrastructure (OCI), bringing together Oracle's global c...
25/08/2026
Siena Kelly, Connor Swindells and Sophie Wilde join Boyle in the eight-part series from Working Title Television and Story Films, directed by Emmy and BAFTA nom...
25/08/2026
The critically acclaimed thriller will return to Sky in NovemberTuesday 25 Augus...
25/08/2026
Wuppertal August 25, 2026
Scottish Rugby Expands Managed Coach Communications Partnership With RiedelRiedel Communications today announced the continued expan...