1. Data Controller
Loris Cramette, individual.
Contact: cramette.loris@gmail.com
2. Data Collected
2.1 User Account
When you sign up, we collect your email address and a hashed password (via Supabase Auth). If you use Google OAuth, we receive your email and Google identifier.
2.2 Usage Data
We collect aggregated daily statistics: number of transcriptions, audio duration, corrections, and LLM tokens consumed. These are synchronized with our servers when you are signed in.
2.3 Audio and Transcription
Audio is captured and transcribed 100% locally on your machine. No audio files are sent to any server. Temporary files are deleted after transcription.
2.4 Corrected Text
In managed mode, selected text is sent to our server which forwards it to OpenAI for correction. We do not store your text content. Correction history is kept locally only.
2.5 Error Reports
We use Sentry to collect crash reports containing stack traces, app version, OS, and an anonymized user identifier (no email).
2.6 Payment Data
Payments are handled by LemonSqueezy. We do not collect or store banking information.
3. Purposes
- Provide the text correction and reformulation service
- Manage your account and subscription
- Track usage quotas
- Improve application stability
4. Legal Basis
Processing is based on contract performance and our legitimate interest (error reports for stability).
5. Data Retention
- Account: retained while active
- Metrics: 12 months
- Error reports: 90 days (Sentry)
- Local history: on your machine, deletable anytime
6. Sub-processors
- Supabase (auth, metrics) — EU
- OpenAI (text correction) — USA
- Sentry (error reports) — EU
- LemonSqueezy (payments) — USA
7. Your Rights (GDPR)
You have the right to access, rectify, delete, port, object to, and restrict processing of your data. Contact: cramette.loris@gmail.com
8. Security
Authentication tokens are encrypted via DPAPI. API keys are encrypted via Electron safeStorage. All communications use HTTPS.