CVE-2026-110698: Out-of-bounds Write in Praat
Description
CVE-2026-110698 is a medium severity vulnerability in Praat version 7.0.02 involving an out-of-bounds write in the whisper_model_load function of the Model Loader component. The issue arises from improper handling of the n_dims argument, which can be manipulated remotely to cause memory corruption. A quick fix has been applied in Praat version 7.0.03, with a more comprehensive fix planned. Exploit code has been publicly disclosed but no known active exploitation in the wild has been reported.
CVSS v4.0
Score 5.3medium
Affected software
Praat
cpe:2.3:a:praat:praat:*:*:*:*:*:*:*:*Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
This vulnerability affects Praat up to version 7.0.02 in the whisper_model_load function located in external/whispercpp/whisper.cpp. The flaw is an out-of-bounds write caused by manipulation of the n_dims argument, allowing remote attackers to potentially corrupt memory. The project maintainer released a quick fix in version 7.0.03 and plans a better fix later. The patch identifier is a910d21a2705ef4cd16d91f29cdc8da1ab1a7221. The CVSS 4.0 base score is 5.3, indicating medium severity with network attack vector, low complexity, no privileges required, and user interaction needed.
Potential Impact
Successful exploitation can lead to out-of-bounds memory write, which may cause application instability, crashes, or potentially allow an attacker to execute arbitrary code or disrupt service remotely. The vulnerability requires no privileges but does require user interaction. The confidentiality, integrity, and availability impacts are low to limited.
Mitigation Recommendations
A patch is available in Praat version 7.0.03 which includes a quick fix for this vulnerability. It is recommended to upgrade from version 7.0.02 to 7.0.03 or later. A more comprehensive fix is planned by the maintainer. Until patched, avoid processing untrusted input that could manipulate the n_dims argument in the whisper_model_load function.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulDB
- Date Reserved
- 2026-10-14T14:13:07.703Z
- Cvss Version
- 4.0
- State
- PUBLISHED
Threat ID: 6acffc4d2cdf04f656b487ec
Added to database: 10/14/2026, 22:03:57 UTC
Last enriched: 10/14/2026, 22:18:24 UTC
Last updated: 10/14/2026, 22:18:24 UTC
Views: 6
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.