Vulnerability · Published May 5, 2025 · Assigned by GitHub_M

CVE-2025-43851GHSL-2025-021_Retrieval-based-Voice-Conversion-WebUI

High 8.9

In plain English

A high-severity unsafe deserialization in Retrieval-based-Voice-Conversion-WebUI from RVC-Project. There are no confirmed reports of exploitation so far. The chance it is exploited in the next 30 days is estimated at 1.0% (higher than 64% of all known vulnerabilities).

Severity (CVSS 4.0)8.9High
Exploit odds (30 days)1.0%top 36.5% of all CVEs
Exploited in the wildNo reportsCISA SSVC: none
Attack toolingNoneno public tooling known

Official description

Retrieval-based-Voice-Conversion-WebUI is a voice changing framework based on VITS. Versions 2.2.231006 and prior are vulnerable to unsafe deserialization. The model_choose variable takes user input (e.g. a path to a model) and passes it to the uvr function in vr.py. In uvr , a new instance of AudioPre class is created with the model_path attribute containing the aformentioned user input. In the AudioPre class, the user input, is used to load the model on that path with torch.load, which can lead to unsafe deserialization and remote code execution. As of time of publication, no known patches exist.

Affected products (as reported)

VendorProductVersions
RVC-Project Retrieval-based-Voice-Conversion-WebUI <= 2.2.231006

Weakness: CWE-502 · Vector: CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:P

References