Approved changes feed: RSS · Atom

cpe:2.3:a:rvc-boss:gpt-sovits:*:*:*:*:*:*:*:*

part: a version: * update: *

VendorRvc Boss (b43499f1-0c6f-57ab-b547-32d42dbf7dc2)
ProductGpt Sovits (1793261d-76fa-5cad-87ea-0dab410954b1)
Edition*
Language*
Software edition*
Target software*
Target hardware*
Other*
NotesImported from purl2cpe mapping

PURL mappings

PURLSourceLast updated
pkg:github/rvc-boss/gpt-sovits purl2cpe 2026-06-01 10:15:30.070708
pkg:pypi/gpt-sovits-python purl2cpe 2026-07-28 16:41:06.919317

Vulnerability references

IdentifiercpeApplicabilitySubmitteddb.gcve.eu detailsRationale
CVE:CVE-2026-63766 vulnerable 2026-07-29 01:06:19.467683 GPT-SoVITS 20250606v2pro OS Command Injection via webui.py
CRITICAL (9.8)
GPT-SoVITS through 20250606v2pro contains an OS command injection vulnerability in webui.py where ASR, slice, denoise, and uvr5 functions interpolate unsanitized Gradio textbox values directly into shell commands executed with shell=True. Attackers can inject shell metacharacters through path parameters to execute arbitrary OS commands as the server process user without authentication.
Published: 2026-07-20T19:18:09.685Z
Updated: 2026-07-28T01:05:53.752Z
Reference links
Imported from gcve-enriched-dumps CVE data
CVE:CVE-2025-49841 vulnerable 2026-06-08 07:29:14.993201 GHSL-2025-053: GPT-SoVITS Deserialization of Untrusted Data vulnerability
GPT-SoVITS-WebUI is a voice conversion and text-to-speech webUI. In versions 20250228v3 and prior, there is an unsafe deserialization vulnerability in process_ckpt.py. The SoVITS_dropdown variable takes user input and passes it to the load_sovits_new function in process_ckpt.py. In load_sovits_new, the user input, here sovits_path is used to load a model with torch.load, leading to unsafe deserialization. At time of publication, no known patched versions are available.
Published: 2025-07-15T20:43:02.861Z
Updated: 2025-07-18T14:42:18.165Z
Reference links
Imported from gcve-enriched-dumps CVE data
CVE:CVE-2025-49840 vulnerable 2026-06-08 07:29:14.992892 GHSL-2025-052: GPT-SoVITS Deserialization of Untrusted Data vulnerability
GPT-SoVITS-WebUI is a voice conversion and text-to-speech webUI. In versions 20250228v3 and prior, there is an unsafe deserialization vulnerability in inference_webui.py. The GPT_dropdown variable takes user input and passes it to the change_gpt_weights function. In change_gpt_weights, the user input, here gpt_path is used to load a model with torch.load, leading to unsafe deserialization. At time of publication, no known patched versions are available.
Published: 2025-07-15T20:42:09.931Z
Updated: 2025-07-16T16:22:55.467Z
Reference links
Imported from gcve-enriched-dumps CVE data
CVE:CVE-2025-49839 vulnerable 2026-06-08 07:29:14.992562 GHSL-2025-051: GPT-SoVITS Deserialization of Untrusted Data vulnerability
GPT-SoVITS-WebUI is a voice conversion and text-to-speech webUI. In versions 20250228v3 and prior, there is an unsafe deserialization vulnerability in bsroformer.py. The model_choose variable takes user input (e.g. a path to a model) and passes it to the uvr function. In uvr, a new instance of Roformer_Loader class is created with the model_path attribute containing the aformentioned user input (here called locally model_name). Note that in this step the .ckpt extension is added to the path. In the Roformer_Loader class, the user input, here called model_path, is used to load the model on that path with torch.load, which can lead to unsafe deserialization. At time of publication, no known patched versions are available.
Published: 2025-07-15T20:40:18.251Z
Updated: 2025-07-16T13:40:13.900Z
Reference links
Imported from gcve-enriched-dumps CVE data
CVE:CVE-2025-49838 vulnerable 2026-06-08 07:29:14.992041 GHSL-2025-050: GPT-SoVITS Deserialization of Untrusted Data vulnerability
GPT-SoVITS-WebUI is a voice conversion and text-to-speech webUI. In versions 20250228v3 and prior, there is an unsafe deserialization vulnerability in vr.py AudioPreDeEcho. The model_choose variable takes user input (e.g. a path to a model) and passes it to the uvr function. In uvr, a new instance of AudioPreDeEcho class is created with the model_path attribute containing the aforementioned user input (here called locally model_name). Note that in this step the .pth extension is added to the path. In the AudioPreDeEcho class, the user input, here called model_path, is used to load the model on that path with torch.load, which can lead to unsafe deserialization. At time of publication, no known patched versions are available.
Published: 2025-07-15T20:36:46.859Z
Updated: 2025-07-16T13:41:00.432Z
Reference links
Imported from gcve-enriched-dumps CVE data
CVE:CVE-2025-49837 vulnerable 2026-06-08 07:29:14.991598 db.gcve.eu details were skipped to keep the page responsive. Imported from gcve-enriched-dumps CVE data
CVE:CVE-2025-49836 vulnerable 2026-06-08 07:29:14.991165 db.gcve.eu details were skipped to keep the page responsive. Imported from gcve-enriched-dumps CVE data
CVE:CVE-2025-49835 vulnerable 2026-06-08 07:29:14.990828 db.gcve.eu details were skipped to keep the page responsive. Imported from gcve-enriched-dumps CVE data
CVE:CVE-2025-49834 vulnerable 2026-06-08 07:29:14.990294 db.gcve.eu details were skipped to keep the page responsive. Imported from gcve-enriched-dumps CVE data
CVE:CVE-2025-49833 vulnerable 2026-06-08 07:29:14.988902 db.gcve.eu details were skipped to keep the page responsive. Imported from gcve-enriched-dumps CVE data

Contribute

You can submit an edit proposal for this CPE entry or suggest a related product/vendor addition using the action button above.