Skip to content
Zahvox
GuideAvailable now

Voice-to-text privacy

Voice recognition is powerful because it sends audio to a large cloud model. That is also the reason to think about what you dictate. Here is the honest picture, without compliance claims.

Try the free voice toolFree · No signup · Runs in your browser

Quick answer

Chrome and Edge send audio to Google, Safari sends audio to Apple, and Zahvox does not store the raw microphone stream. Do not dictate regulated, personal, or confidential data into any browser recognizer.

How browser voice-to-text works

The Web Speech API opens the microphone in the browser and streams audio to a cloud recognizer. The vendor returns text and the browser passes that text to the page. The website itself does not touch the audio unless it also records it separately.

What Zahvox does today

  • The free tool at /tool listens to the browser's recognizer, not your raw mic.
  • The transcript stays in the browser tab session.
  • There is no account, no cloud storage, and no team database yet.
  • Saved history, personas, and templates are planned and will be opt-in once launched.

What Zahvox does not do

  • It does not upload raw audio to Zahvox servers.
  • It does not sell or share transcripts.
  • It does not run background recording between sessions.

When voice-to-text is the wrong choice

  • Patient identifiers, medical history, or clinical notes for regulated environments.
  • Legal case detail with client confidentiality.
  • Finance data covered by internal handling policies.
  • Passwords, security answers, or any credential.

Practical habits that reduce exposure

  • Dictate on a personal device, not a shared corporate one, for personal writing.
  • Redact names and identifiers before dictating; put a placeholder like Client A.
  • Close the tab when you are done so the browser drops the session.
  • Clear browser storage if you shared the device.

Where to read the specifics

The privacy policy covers what data touches Zahvox. The data handling policy explains how transcripts and account fields will be handled once accounts exist. The security page lists the boundaries we hold today.

Frequently asked questions

Keep exploring

Browser voice-to-text (no download)

How the browser-only workflow works.

Voice-to-text not working

Troubleshoot if the recognizer fails silently.

Web Speech API

The underlying browser feature.

Privacy policy

What we do and do not collect.

Data handling policy

How future account data will be handled.

Security

Zahvox's current security posture.

Free voice tool

Open the tool with the privacy defaults.

Stop typing. Start thinking.

Open the free voice-to-text tool and turn your next thought into clean text in seconds.

Open the free tool