SpeechDone

Subprocessors

3. Annex 2: Subprocessors

As of October 2026. We announce changes 30 days in advance (§ 6).

SubprocessorRegistered officePlace of processingTaskDataWhenThird-country basis
Hetzner Online GmbHGunzenhausen, GermanyGermanyServer, data center, off-site backupall data on the serveralwaysnot applicable
Cloudflare, Inc.San Francisco, USAglobal network, connection to the nearest locationNetwork, protection against attackstraffic between app and server (forwarded encrypted)whenever the server is contactedEU-U.S. Data Privacy Framework, Standard Contractual Clauses
Deepgram, Inc.San Francisco, USAEU serversSpeech recognition (text while speaking)voice recordingfor every recognition through our serverprocessing in the EU; EU-U.S. Data Privacy Framework for access from the USA
Mistral AI SASParis, FranceEU serversSpeech recognition (inserted text), rewriting, editing, translating, questions, style analysisvoice recording, textsfor recognition and these features through our servernot applicable
Resend (Plus Five Five, Inc.)San Francisco, USAEU (Ireland)Sending of account emailsemail address, content of the account emailfor account emailsEU-U.S. Data Privacy Framework
Functional Software, Inc. (Sentry)San Francisco, USAdata region FrankfurtCrash reports of the apptechnical data without name, email, IP and without dictation contentApp Store versionprocessing in the EU; EU-U.S. Data Privacy Framework for access from the USA

Apple (Sign in with Apple, iCloud, notifications, App Store) is not a subprocessor: the user uses these services in the user’s own contractual relationship with Apple.

For providers based in the USA, transfers rely on the EU-U.S. Data Privacy Framework (Art. 45 GDPR) where the provider is certified. These are Cloudflare, Deepgram, Resend and Sentry. Deepgram, Resend and Sentry process on servers in the EU; the Data Privacy Framework covers possible access from the USA.