Privacy Policy

VoiceFlow is built as a local-first Windows assistant.

This policy explains what VoiceFlow V3 stores on your PC, what is sent to AI providers when you choose AI features, and how website feedback is handled.

Short version

VoiceFlow does not continuously watch your computer. Dictation audio is processed with the bundled local speech model. Local history, screenshots, settings, and diagnostics stay on your Windows PC unless you choose to use an AI feature that sends selected content to an external AI provider.

Who operates VoiceFlow

VoiceFlow is an independent Windows desktop app created by Naveen Pothu. For privacy questions, support, or deletion requests related to website feedback, contact naveenpothu2728@gmail.com.

Information stored locally

VoiceFlow V3 may store app data on your Windows PC so the assistant can work smoothly.

  • Recent generated results and history that you can copy again from the orb menu.
  • Screenshots or cropped screenshots you choose to capture or attach.
  • Settings such as microphone choice, hotkey behavior, speech options, startup preferences, and app preferences.
  • Diagnostics information such as API key status, AI connection status, speech model status, microphone status, hotkey status, and the last error.

You can clear history from inside the app. Removing the app and its local data from your PC removes locally stored VoiceFlow data.

Voice dictation and audio

Voice dictation is designed to run locally using the speech model bundled with the Windows build. When you hold Right Shift, speak, and release, VoiceFlow transcribes the recording on your computer and pastes the resulting text into the active app.

VoiceFlow does not upload dictation audio for the local transcription feature. If a future build adds optional cloud speech features, they should be clearly presented as optional before use.

AI text cleanup and commands

AI features require your own Groq API key. When you use AI cleanup, AI chat, voice commands, clipboard rewrites, or command-style prompts, VoiceFlow may send the selected text, dictated prompt, chat message, or copied content needed for that request to Groq.

VoiceFlow should not send everything on your computer. It sends the content needed for the specific AI action you choose.

Screenshot and ask screen

VoiceFlow V3 can capture the current screen, let you crop it, attach the cropped image to a chat message, and ask questions such as "What is on my screen?", "Explain this error", or "What should I click next?"

Only the screenshot or cropped screenshot you select for that AI request should be sent to the vision AI provider. Avoid sending screenshots that contain passwords, private messages, financial information, medical information, API keys, or other sensitive data.

API keys

AI features require a user-provided Groq API key. Do not share your API key with other people, and rotate it from the provider dashboard if you believe it was exposed.

Website feedback

The VoiceFlow website does not use a third-party tracking tool. It does not track page views, scroll depth, section views, or download clicks through an external dashboard.

If you submit the feedback form, the website receives the name, email address, and feedback text you enter through the website API. Do not include passwords, API keys, private documents, or other sensitive information in feedback.

What VoiceFlow does not do

  • VoiceFlow does not require a VoiceFlow account.
  • VoiceFlow does not continuously upload your desktop activity.
  • VoiceFlow does not upload local dictation audio for the bundled offline speech model.
  • VoiceFlow does not send screenshots unless you choose a screenshot-based AI request.
  • VoiceFlow does not intentionally collect payment details through the app or website.

Third-party services

VoiceFlow and its website may use third-party services, including GitHub for downloads and Groq for AI features when configured. Each third-party service may process data according to its own privacy policy.

Security and beta status

VoiceFlow V3 is a beta Windows app. Windows may show an "Unknown publisher" warning until the app is code-signed. The app includes diagnostics so users can inspect connection, microphone, hotkey, speech model, API key, and error status.

Because VoiceFlow is still evolving, avoid using it with highly sensitive information unless you are comfortable with the selected AI provider receiving the text or screenshot you choose to send.

Children's privacy

VoiceFlow is not designed for children under 13. If you believe a child has submitted personal information through the website feedback form, contact naveenpothu2728@gmail.com so it can be reviewed.

Changes to this policy

This privacy policy may be updated as VoiceFlow adds features, changes providers, or improves data handling. The latest version will be available from the Privacy Policy link on the VoiceFlow website.

Last updated: June 25, 2026

Contact

Questions, feedback, or privacy requests can be sent to naveenpothu2728@gmail.com. You can also connect through the LinkedIn link on the VoiceFlow website.