Tryggleiki og persónsupplýsingar
Nógvar av teimum eru ókendur, men tað er ikki ókent, hvussu teir hava virkað.
Three Levels of Privacy
Standard
All users, all plans — including free
- HTTPS (TLS 1.3) fyri øll data í transit
- Lyd slettað beinanvegin eftir viðgerð
- Listi yvir søguligar bygdir í Føroyum
- We can read stored transcripts
- Data never sold or used for training
- Slepp tínum upplýsingum tá sum helst
Private transcript
Pro+ Til ber at keypa Pro og Business planir.
- All in Standard, plus:
- Skriftin er krypterað í tínum vafra (AES-256-GCM) áðrenn hon verður goymd
- We store only encrypted data — we can't read it
- Key derived from your password, never sent to us
- ⚠ Audio is still being processed on our servers during transcription
Private Cloud / Self-Hosted
Full isolation — frá $99/mo
- Lyd fer ongantíð úr tínum innrætting
- Umseting koyrir á tínum GPU
- No data sent to STT.ai servers
- Air-gapped support available
- End-to-end privacy
What Actually Happens to Your Data
Ein lýsing av, hvussu skipanin er skipað og virkar.
Standard (all users)
1
Tú sendir upp ljóð ella tekur upp beinleiðis
Tí er tað ikki neyðugt at brúka HTTPS (TLS 1.3) til at senda boð til ein GPU-tølvara.
2
Lyd verður arbeitt við í minninum
Våre AI-modeller transskriberer dit lyd på GPU'en. Lyden holdes i hukommelsen under bearbejdning - aldrig skrevet til disk - og slettes fra hukommelsen straks efter.
3
Texturin er lagdur í okkara database
Text transcript, timestamps, and speaker labels are saved so you can access them later. We can read these data (this is how search, AI summaries, and sharing work).
4
You can delete everything whenever
Slepp einstøkum útskriftum ella heili kontoin tín úr Privacy Settings. Sleppingin er varandi og skjót. Innstillingar fyri persónsupplýsingar
With Private Transcript enabled
Steps 1-2 are the same — your audio must be processed on our servers to generate the transcript. The difference is what happens next:
3
Textin verður krypteraður í vafranum, áðrenn hann verður lagdur
Eftir at hava skrivað niður, verður úrslitið sent aftur til vafrann. Vafrann krypterar tað við AES-256-GCM við at brúka ein lykil, sum er útleiddur úr tínum loyniorði (PBKDF2, 100K iterations). Den krypterede blob sendes derefter til vores servere til lagring. Vi ser eller gemmer aldrig krypteringsnøglen.
4
We only store encrypted data
Databásurin okkara inniheldur bert tað krypteraða blob'ið. Vit kunnu ikki dekryptera tað. Um okkara databásur varð brotið, so verða útskriftirnar ikki lesandi.
Viktigt: Private transcript verndar geymda transkriptiónina. Áðrenn transkriptiónin fer fram, verður ljóðið arbeitt við á okkara netverkum fyri at gera tekstin. Um títt hóttanarmodell krevur, at ljóð ikki snertir 3. partar, kanst tú hugsa um Private Cloud ella Self-Hosted. Private Cloud
What We Can and Can't See
We CANNOT see (with Private Transcript)
- Tín goymdi transkriptiónstexta
- Name of speaker or label (stored)
- Time stamps or word-level data (stored)
- Títt krypteringsnøkl ella passorð
We CAN see (even with Private Transcript)
- Lydtøkan undir viðgerð (sleppt eftir)
- File name, size, duration (metadata)
- Spælið er funnið, modellið er brúkt
- Time stamp of transcription
- Tínar kontoupplýsingar og fakturering
Tekniskar upplýsingar
| Algorithm | AES-256-GCM (authenticated encryption) |
| Key derivation | PBKDF2 with SHA-256, 100,000 iterations |
| IV (nonce) | Tilvildarlig 12 bytes pr. kryptering (aldrei nýtt aftur) |
| Key storage | Aldrei goymt - kemur frá loyniorðini í hvørjari setu |
| Transportkryptering | TLS 1.3 (HTTPS) + HSTS (1 ár, preload) |
| Audio retention | Processed in memory, never written to disk, deleted immediately |
| Implementering | Web Crypto API (browser-native, no external libraries) |
| Source code | github.com/sttaigit/stt-encryption (Mynd: |
Private Transkription Trade-offs
Private transcript is opt-in because encrypting the stored transcript limits some features:
Virkar við kryptering
- Viewing your transcripts
- Exporting (TXT, SRT, VTT, etc.)
- Niðurlating
- Redigering (decrypted in browser)
Ikki tiltækt við kryptering
- Søk á server-side yvir transkriptiónir
- AI summaries and chat (server can't read data)
- Public sharing via link
- Team Workspace Collaboration
Men tað er ikki altíð, at fólk hava møguleika at lesa.
Private transcript verndar transkriptiónina í hvíld, men ljóð fer framvegis ígjøgnum GPU'ina meðan tað verður arbeitt við. Um tú hevur eftirlits- ella sikringskrav, sum krevja, at ljóð ikki snertir infrastruktur frá øðrum, so eru hesi valmøguleikarnir:
Private Cloud
$499/mánað
Dedikeraður GPU-tølvari, ið verður umsitin av okkum. Lydið fer ongantíð úr tínum isoleraða umhvørvi.
- 100 f.Kr.
- Isolated — no shared infrastructure
- Lyd varð bert arbeitt við á tínum maskinvara
- Full API access + SLA
Self-Hosted
$99/mánað
Docker mynd. Tínar serverar. Tín GPU. Innihaldið fer ikki út um tín netverksstól.
- Docker - koyrir á hvørjum NVIDIA GPU
- Air-gapped support — no internet required
- Model updates included
- Full control, full privacy
Våre forpligtelser (Alle brugere, alle planer)
- Síðani verður lagið útgivið í 24 londum. Processed in GPU memory, source audio removed shortly after transcription. The only exception: if you choose "Contribute corrections + audio to Voice Lab" in Privacy Settings, audio is archived for up to 90 days while the corrections-ingest cron extracts clipped segments under CC-BY-SA-4.0. Innstillingar fyri persónsupplýsingar
- Dataini verða ongantíð brúkt til AI-trening um tú ikki úttrykkiliga velur tað í Privacy Settings (text-only corrections training and/or Voice Lab audio contributions — both default off). Innstillingar fyri persónsupplýsingar
- Tað eru ikki øll, sum hava data. Aldrig, aldrig, aldrig.
- All traffic encrypted in transit 1.300 íbúgvar.
- Slepp tínum upplýsingum tá sum helst frá Privacy Settings ella við at sletta tín konto. Innstillingar fyri persónsupplýsingar
- Open source software — audit it yourself (MIT license). audit it yourself
Open Source Cryptography
Krypteringssavnið er fullkomiliga opið undir MIT-leyvinum. Treystið okkum ikki - kannið kódina. Engin treyt um at trúgva, bara matematikk.
Sí miðlasavnið »GitHub« í Wikimedia Commons. | Sýna keldaErt tú klárur at skriva trygt?
Upload your first file free. Private transcripts available on Pro and Business plans.
Start TranscribingOfta settir spurningar
STT.ai security and encryption runs in your browser: paste a URL, upload a file, or record from your mic. STT.ai picks the AI model and returns the transcript in under 5 minutes. Export as TXT, SRT, VTT, DOCX, JSON, or PDF.
Yes — every visitor gets 600 free minutes/month on STT.ai, usable for STT.ai security and encryption the same as any other workflow. Paid plans starting at $5/month unlock longer files, private transcripts, and priority queueing.
STT.ai security and encryption runs on the same AI models as the rest of STT.ai — our best models reach 95-97% accuracy on clean speech (3-5% Word Error Rate on benchmarks). Switch models on the fly if the first pass is below your target.
STT.ai security and encryption can run on any of STT.ai's 10+ models — STT.ai Enhanced (most accurate), Whisper Large V3 (99 languages), NVIDIA Canary (#1 WER on supported langs), Whisper Turbo (fast), Moonshine (lightweight), and more.
Yes. Every transcript exports as SRT or VTT — works with YouTube, Vimeo, TikTok, VLC, and every major video player. The burn-subtitles tool overlays them onto video as hardsubs.
Yes. Speaker diarization automatically labels each voice (Speaker 1, Speaker 2, ...) and you can rename them in the built-in editor. Works across all models and languages.
Most STT.ai security and encryption jobs finish in under 5 minutes. A 1-hour audio file typically completes in 2-3 minutes with our fastest models. Speed depends on chosen model and current GPU load.
STT.ai security and encryption accepts 20+ formats — MP3, WAV, M4A, FLAC, OGG, MP4, MKV, MOV, WebM, AVI, and more. Output to TXT, SRT, VTT, DOCX, JSON, or PDF.
Yes. Audio files submitted to STT.ai security and encryption are processed and deleted by default. Pro plans add client-side encryption — even if STT.ai's database is breached, your transcripts are unreadable without your key. Data is never used for model training without explicit opt-in.
Yes. STT.ai offers a REST API with Python and Node.js SDKs, plus an MCP server for Claude and Cursor — all usable for STT.ai security and encryption workflows. Free API tier includes 100 minutes/month.
Yes. Every transcript opens in the built-in editor where you can correct words, rename speakers, adjust timestamps, and add notes. All changes save automatically.
Every transcript gets a unique shareable URL. Export to DOCX or PDF for email. Pro plans add password-protected and permanent links — useful for client work.
STT.ai handles 1,300+ platforms including YouTube, Vimeo, TikTok, SoundCloud, Zoom, Google Meet, podcast hosts, and more. URL transcription works with publicly-available content only — DRM-protected sources can't be transcribed.