← Back to PrepCall

Voice Data Retention and Destruction Policy

Last updated: 18 August 2026 (version 1.1)

This policy sets out how PrepCall retains and destroys voice data, including any data that may be treated as a biometric identifier or biometric information under laws such as the Illinois Biometric Information Privacy Act, the Texas Capture or Use of Biometric Identifier Act, and equivalent provisions elsewhere. It sits alongside our Privacy Notice, which covers all other personal data.

1. What we hold

During a practice session we process the audio of your speech in real time. From that we produce two things:

  • a written transcript of what you said; and
  • numerical measurements of how you sounded: headline delivery scores such as pace, energy and steadiness, and the detailed moment-by-moment vocal expression measurements produced by the analysis to support them.

All of these are measurements of how you sounded. They are not a measurement of what you felt, and they are not a clinical or psychological assessment.

2. What we do not hold

We do not store the audio recording of a practice session. Audio is processed in real time by our processor, Hume AI, Inc., which is configured not to retain it. No recording of your voice exists after your session ends, with us or with our processor.

We do not create, store or use a voiceprint, speaker embedding or any other template capable of identifying you from your voice. We do not use voice data to identify anyone.

3. Purpose

Voice data is processed for one purpose only: producing your own coaching report, and tracking your own progress across your own sessions. It is used for nothing else.

4. Retention

Transcripts and the vocal measurements described above are retained for 12 months from your most recent practice session.

They are deleted sooner if any of the following happens:

  • you delete your account;
  • you withdraw your consent to voice processing; or
  • you request deletion.

5. Destruction

Records that pass the retention period are permanently deleted from our production database, and from backups on the next backup rotation. Destruction is permanent and irreversible. We do not retain archival copies.

6. Disclosure

We do not sell, lease, trade or otherwise profit from voice data.

We disclose session audio to Hume AI, Inc. only for the purpose of performing the analysis described above, and only with your separate consent.

We do not disclose voice data, transcripts or vocal measurements to employers, recruiters, hiring platforms, advertisers or data brokers. PrepCall is not sold to organisations for the purpose of assessing candidates or employees, and its output does not feed any hiring decision.

7. Your requests

You can delete your voice data at any time by deleting your account (open the account menu, then Manage account, then Delete account), which permanently removes your transcripts, vocal measurements and reports. You can also email privacy@prepcall.me and we will delete it for you. We respond to email requests within 30 days.

8. Review

This policy is reviewed at least annually, and whenever our voice processing or our processors change.

Version 1.1 — 18 August 2026 — clarified that stored vocal measurements include the detailed moment-by-moment expression measurements, not only headline delivery scores. Version 1.0 — August 2026 — first publication.