Skip to content

Security: owenawsong/Inflect

Security

SECURITY.md

Security Policy

Inflect is a speech-generation research project. Security issues may involve code execution, model artifacts, data handling, or misuse of voice-cloning workflows.

Reporting

If you find a security issue, do not open a public issue with exploit details.

Send a private report to the repository owner through GitHub, or open a minimal private advisory if available.

Include:

  • affected files or workflow
  • reproduction steps
  • impact
  • suggested fix if known

Voice and Data Safety

Do not submit:

  • private voice recordings
  • credentials
  • Hugging Face tokens
  • API keys
  • copyrighted datasets without clear rights
  • model checkpoints containing private training data

Supported Status

This repo is currently a research preview. There is no production support guarantee yet, but credible vulnerability reports will be treated seriously.

Responsible Use

Inflect is intended for consent-based voice work, local creative tooling, accessibility, research, and product prototyping.

Do not use it for impersonation, fraud, harassment, or non-consensual voice cloning.

There aren't any published security advisories