Your voice. Your data.
Your control.
We built Swiftalk with privacy as a core principle, not an afterthought. You choose what data leaves your device, if any.
You Choose Your Privacy Level
Local processing for complete privacy, cloud for maximum accuracy
Local Processing
Maximum privacy
- Audio stays on your device
- Works completely offline
- No data sent to servers
- ~95% accuracy (English)
- Requires local model download
Cloud Processing
Maximum accuracy
- Audio encrypted in transit
- Processed in secure data centers
- Deleted immediately after processing
- ~99% accuracy (all languages)
- Requires internet connection
You can switch between modes at any time, even on a per-recording basis.
Security Practices
Industry-standard security measures to protect your data
Encryption
All data encrypted at rest (AES-256) and in transit (TLS 1.3). Your audio and transcripts are protected at every step.
Data Storage
Local recordings stay on your device. Cloud transcriptions are processed and deleted within minutes, never stored permanently.
Privacy First
We don't sell, share, or use your data for training. What you say stays between you and your computer.
Access Control
Your account is protected with secure authentication. Optional two-factor authentication for enterprise users.
Security Updates
Regular security patches and updates. Automatic updates keep you protected without any action required.
Compliance
Built with GDPR compliance in mind. Data processing agreements available for enterprise customers.
What Happens to Your Audio
A transparent look at our data flow
You speak
Audio captured by your microphone
Based on your settings
Local
Processed on device
Never leaves your computer
Cloud
Encrypted in transit
Deleted after processing
Transcript delivered
Text appears at your cursor
Why We Need Each Permission
Complete transparency about what we access and why
Microphone
RequiredRequired to capture your voice for transcription. Audio is only recorded when you activate Swiftalk.
Accessibility / Keyboard
RequiredRequired for global hotkey functionality and to type transcribed text into any application.
File System
RequiredUsed to store your recordings locally, save settings, and manage the local AI model files.
Network
Required for cloud transcription, syncing settings, checking for updates, and license verification.
Third-Party Services
Transparent disclosure of our partners and what data they receive
| Service | Purpose | Data Shared | Retention |
|---|---|---|---|
| OpenAI | Cloud transcription (Whisper API) | Audio (when using cloud mode) | Deleted after processing |
| Polar | Payment processing | Email, payment method | As required for billing |
| Better Auth | Authentication | Email, OAuth tokens | Session duration |
We carefully vet all third-party services and only share the minimum data required.
Report a Security Vulnerability
We take security seriously. If you've discovered a vulnerability, please report it responsibly. We commit to responding within 24 hours.
Ready to get started?
Experience voice dictation that respects your privacy.