gregdolanQ: User privacy
Hi Voicedash Team,
I’m interested in your service but need to clarify your Privacy Policy:
Transcription vs. Email: Your policy mentions "email content." As this is a voice tool, does this refer to my transcripts? Please confirm this is just a template error.
Audio Storage: Do you store raw audio recordings on your servers? Under GDPR, voice data is sensitive. Is audio deleted immediately after processing?
Training & Retention: You state data isn't used for training. To confirm, is this because you use the OpenAI API? How long is the transcript text retained on your servers before being purged?
Data Rights: How do I exercise my GDPR "Right to Erasure" to permanently delete my account and all associated metadata?
emad.puursina
Edited Mar 16, 2026A: Hello,
Thank you for reaching out with your detailed privacy questions. I am happy to clarify these points for you.
Transcription vs. Email: Could you please specify exactly where you saw the term "email content"? I want to ensure I address the correct part of the document. To clarify our process: we do not store any audio recordings or transcriptions on our servers. Your audio is sent directly to the OpenAI API for processing, and we receive the resulting text transcript back to display in your application. We do not use your data for any training purposes.
Account Deletion & Data Rights: To exercise your "Right to Erasure," you can simply send us a deletion request by replying to this email or contacting us at [email protected]. We will process your request and permanently delete your account and all associated metadata from our systems.
Please let me know if you have any other questions.
I still have questions about data security.
🤔
Is it possible to choose an LLM model for processing my data?
Is it possible to use BYOK for sensitive data?
Have you considered using local models directly on my computer?
At the moment, choosing an LLM modeli is not possible for processing your data.
BYOK is not available yet, but it is planned for a future release for Tier 2 and above. We’re working toward adding this option for users who need additional control over sensitive data.
Running local models directly on your computer is not currently supported.
When might this happen?