
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 ...
04/08/2026
Dalet, a leading technology and service provider for media-rich organizations, t...
04/07/2026
April 7 2026, 19:00 (PDT) Detective Conan: Fallen Angel of the Highway Opens in...
01/06/2026
January 6 2026, 05:30 (PST) Dolby Sets the New Standard for Premium Entertainment at CES 2026
Throughout the week, Dolby brings to life the latest innovatio...
02/05/2026
Dalet, a leading technology and service provider for media-rich organizations, t...
01/05/2026
January 5 2026, 18:30 (PST) NBCUniversal's Peacock to Be First Streamer to ...
21/04/2026
MAS and Lockheed Martin partner to establish an F-35 depot in Canada, enabling in-country sustainment and creating high-skilled aerospace jobs....
21/04/2026
Advertising strategies shift as competition grows for a large, active and qualit...
21/04/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
21/04/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
21/04/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
21/04/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
21/04/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
21/04/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
21/04/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
21/04/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
21/04/2026
Layercake Deepens Bitmovin Integration to Power End-to-End Media Orchestration w...
21/04/2026
Cobalt Digital Announces Launch of blueCORE at NAB Show 2026: Compact Powerhouse...
21/04/2026
On-Premise AI Suite from Studio Network Solutions Debuts at NAB Show 2026
Melanie Ciotti April 21, 2026
0 Comments
Unlimited processing, no cloud depe...
21/04/2026
London, 21 April 2026 IBC today announced the appointment of Tim Banham as its first Chief Commercial Officer (CCO), a newly created role that reflects the or...
21/04/2026
Motion Design Tools - April 2026
Roland Kahlenberg April 21, 2026
0 Comments
Within 2 days, Maxon and Canva announced pro-level motion design apps - A...
21/04/2026
Chaos and Zero Density to Showcase Real-Time Ray Tracing for Virtual Studios and...
21/04/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
21/04/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
21/04/2026
Cinema 4D brings professional 3D workflows to iPad. The return of Autograph now free for individual users. ZBrush expands to Windows on Arm. See it all at NAB...
21/04/2026
Software version 1.6 extends enterprise functionality to place Buttons at the heart of media operations at any scale
Bitfocus, the Norwegian software develope...
21/04/2026
Cobalt Digital Announces Launch of blueCORE at NAB Show 2026: Compact Powerhouse Signal Processors for SDI and ST 2110 Workflows
Compact, multi-function stan...
21/04/2026
Applications open for 2026 AISF and Screen Australia Writer/Director Virtual Ses...
21/04/2026
Comscore Continues Building Momentum in Local Measurement Through New Agreements...
21/04/2026
Summer is nearly here and Super Garden is returning to our screens to spark some gardening inspiration. The new series kicks off on Thursday 23 April at 7pm on ...
20/04/2026
At the 2026 NAB Show, Sony is showcasing a broad slate of innovations across liv...
20/04/2026
At the 2026 NAB Show, Canon is doubling down on its commitment to live sports pr...
20/04/2026
Fujifilm is sharpening its focus on core broadcast production with a new wave of...
20/04/2026
This upcoming summer in North America is going to be a busy one. The 2026 FIFA M...
20/04/2026
Glookast (Booth W1661) announced a series of product updates at NAB Show 2026, c...
20/04/2026
Matrox Video and Amagi announced a collaboration to integrate the Matrox ORIGIN ...
20/04/2026
Riedel Communications (Booth C4908) announced that the Asociaci n del F tbol Arg...
20/04/2026
Ikegami (Booth C3819) announced the VFE-P07D monocular OLED viewfinder at NAB Sh...
20/04/2026
International Association of MediaTech (IAMT), formerly known as IABM, announced...
20/04/2026
Harmonic (Booth W2831) announced that DIRECTV is updating its US direct-to-home (DTH) video platform using Harmonic's VOS Media Software.
The deployment is...
20/04/2026
Wasabi Technologies announced that it has acquired the Lyve Cloud business from Seagate Technology. As part of the agreement, Seagate received equity in Wasabi ...
20/04/2026
EVS (Booth N1841) has launched Choreon, a robotics controller for media producti...
20/04/2026
The NAB Show is in full swing, and the SVG and SVG Europe editorial teams are chasing down the hottest stories from all over the Las Vegas Convention Center. He...
20/04/2026
Skyline Communications announced the availability of its DataMiner xOps platform...
20/04/2026
Studio Network Solutions (Booth N1129) introduced a set of new products at NAB S...
20/04/2026
Dell Technologies is showcasing its Dell AI Data Platform with NVIDIA at NAB Sho...
20/04/2026
Blackmagic Design has announced Fairlight Live, a software-based live audio mixer with SMPTE 2110 support and spatial audio mixing. A public beta is available n...
20/04/2026
At the 2026 NAB Show in Las Vegas, Imagine Communications VP of Sales, Sports an...
20/04/2026
At the 2026 NAB Show in Las Vegas, LiveU Senior Director of Sales, Sports Philli...
20/04/2026
A song that perfectly captures a moment is magic. But when you uncover the story behind it, who made it, what inspired it, and the meaning woven into the lyrics...