Keamanan & Privasi
Apa kang kelakon karo audio lan transkripsi ing saben langkah. Ora ana marketing fluff - mung cara kerjane.
Tiga Tingkat Privasi
Standar
Saben pangguna, saben rencana — kalebu gratis
- HTTPS (TLS 1.3) kanggo kabèh data ing transit
- Audio dihapus langsung sawisé diproses
- Transkrip sing disimpan ing basis data kita
- Kita bisa maca transkripsi sing disimpan
- Data ora pernah didol utawa digunakake kanggo latihan
- Hapus data sampeyan kapan wae
Transkripsi Pribadhi
Pro+ Digunakaké ing Pro lan Business plans
- Kabeh ing Standar, ditambah:
- Transkripsi kang dienkripsi ing panyelehan (AES-256-GCM) sadurunge disimpan
- Kita mung nyimpen data sing dienkripsi — kita ora bisa maca
- Kunci dijupuk saka sandi sampeyan, ora diutus marang kita
- ⚠ Audio isih diproses ing penjelajah kita nalika transkripsi
Cloud pribadi / Self-hosted
Full isolation - saka $99/mo
- Audio ora tau ninggalake infrastrukturmu
- Transkripsi diwiwiti ing GPU sampeyan
- Ora ana data kang dikirim menyang STT.ai penjelajah
- Dukungan Air-gapped ana
- Privasi pungkasan-nganti-akhir
Apa sing kedadeyan karo datamu
Transparan, langkah-demi-langkah breakdown saka carane audio lan transkripsi sampeyan ditangani.
Standar (kabeh pangguna)
1
Sampeyan nyathet audio utawa rekaman langsung
Fail sampeyan dikirim liwat HTTPS (TLS 1.3) menyang server GPU kita kanggo transkripsi.
2
Audio diproses ing ingged
Model AI kita ngrekam audio ing GPU. Audio iku disimpen ing memori nalika proses — ora tau ditulis ing disk — lan dipasak saka memori langsung sawisé.
3
Transkrip wis disimpan ing databasa kita
Transkrip teks, timestamps, lan label juru bicara disimpan supaya sampeyan bisa ngakses banjur. Kita bisa maca data iki (iki cara panelusuran, ringkasan AI, lan kerja bagéan).
4
Sampeyan bisa mbusak kabeh kapan wae
Pa_cahaké transkripsi individu utawa kabèh akunmu saka Tetep Privasi. Pa_cahaké iku permanen lan langsung. Tetepkan Privasi
Kanthi Privasi Transcript diaktifake
Langkah 1-2 padha — audiomu kudu diproses ing penjelajah kita kanggo ngasilake transkripsi. Perbedaan iku apa kang kelakon sabanjuré:
3
Transkripsi dienkripsi ing panyelehanmu sadurunge disimpan
Sawisé transkripsi, asil bakal dikembalikaké menyang panyelenggu. Panyelenggu bakal nyenkripsi kanthi AES-256-GCM nganggo kunci kang dijupuk saka tembung sandhi (PBKDF2, 100K iterasi). Blob sing dienkripsi banjur dikirim menyang panyelenggu kanggo disimpan. Kita ora bakal ndeleng utawa nyimpen kunci enkripsi.
4
Kita mung nyimpen data sing dienkripsi
Panggonan data basis kita mung ngandhut blob sing dienkripsi. Kita ora bisa nyaringé. Yen panggonan data basis kita dirusak, transkripsi sampeyan ora bisa dibaca.
Penting: Transkripsi Pribadhi nglindhungi transkripsi sing disimpan. Nalika proses transkripsi, audio sampeyan diproses ing server kita kanggo ngasilake teks. Yen model ancaman sampeyan mbutuhake audio ora tau nyentuh server pihak katelu, pikirake Cloud Pribadhi utawa Self-Hosted. Privasi
Apa kang bisa lan ora bisa dideleng
Kita NAKENE ndeleng (kaliyan Transkripsi Pribadi)
- Teks transkripsi sing disimpan
- Nama utawa label pamicara (disimpen)
- Timestamp utawa data tingkat tembung (disimpen)
- Sandi utawa kunci enkripsi sampeyan
Kita KANGGUH ndeleng (utawa nganggo Transcript Pribadhi)
- Audio sampeyan nalika diproses (dihapus sawisé)
- Nama berkas, ukuran, durasi (metadata)
- Basa ditemokaké, model kang digunakake
- Transkripsi
- Informasi akun lan mbayarmu
Rincian teknis
| Algoritme enkripsi | AES-256-GCM (penyulitan kang diverifikasi) |
| Derivasi kunci | PBKDF2 karo SHA-256, 100,000 iterasi |
| IV (nonce) | 12 bait kanggé saben enkripsi (ora pernah dipungunakaken malih) |
| Pendhidhikan | Ora disimpen — dijupuk saka sandi ing saben sesi |
| Enkripsi Transportasi | TLS 1.3 (HTTPS) + HSTS (1 taun, preload) |
| Retention audio | Diproses ing ing ing, ora tau ditulis ing disk, dipasak langsung |
| Implementasi | Web Crypto API (browser-native, ora ana pustaka njaba) |
| Kode sumber | github.com/sttaigit/stt-encryption (Lisensi MIT) |
Transkrip Privasi
Transkripsi pribadi iku opt-in amarga enkripsi transkripsi sing disimpan nglibatake sawetara fitur:
Nggunakaké enkripsi
- Nglakoni transkrip
- Eksportasi (TXT, SRT, VTT, lsp.)
- Muter-muter
- Ngedit (didekkripsi ing panyelehan)
Ora ana karo enkripsi
- Pencarian server-side liwat transcripts
- Ringkasan AI lan chatting (panggina ora bisa maca data)
- Berbagi umum liwat tautan
- Kollaborasi ruangkerja tim
Perlu audio kanggo ora ninggalake server sampeyan?
Transkripsi pribadhi nglindhungi transkripsi nalika ora digunakaké, nanging audio isih bisa liwat GPU nalika diproses. Yen syarat keamanan utawa kepatuhan sampeyan mbutuhake audio ora bisa nyawang infrastruktur pihak katelu, iki pilihanmu:
Privasi
$499/wulan
Server GPU khusus kang diurus déning kita. Audiomu ora tau ninggal lingkungan sing diisolasi.
- A100 GPU sing didedikasikaké
- Isolasi - ora ana infrastruktur dipérang
- Audio mung diproses ing piranti kerasmu
- Full API akses + SLA
Self-hosted
$99/wulan
Docker image. Servermu. GPUmu. Ora ana sing ninggal jaringanmu.
- Docker — dioperasikaké ing GPU NVIDIA apa wae
- Dukungan Air-gapped - ora perlu internet
- Model anyar kalebu
- Kontrol lengkap, privasi lengkap
Kita Kawajiban (Sedaya Pengguna, Sedaya Plan)
- File audio dipunpasak ing 24 jam kanthi default. Diproses ing memori GPU, sumber audio dicopot sadurungé transkripsi. Saben-saben eksepsi: yen sampeyan milih "Contribute corrections + audio to Voice Lab" ing Pengaturan Privasi, audio bakal diarsip nganti 90 dina nalika cron-ingest ngresiki segmentasi sing dipotong ing CC-BY-SA-4.0. Tetepkan Privasi
- Datamu ora tau digunakake kanggo latihan AI kajaba sampeyan milih ing Setelan Privasi (latihan pambenahan teks-saben lan/utawa kontribusi audio Voice Lab — kaloro diaktifaké kanthi default). Tetepkan Privasi
- Kita ora adol datamu. Kabeh.
- Sedaya lalu lintas dienkripsi nalika transit Nokia 1300 dipirantèni kodhak 1.3 MP.
- Hapus data sampeyan kapan wae saka Setelan Privasi utawa kanthi mbusak akunmu. Tetepkan Privasi
- Kode enkripsi punika sumber-buka — audit iku dhewe (MIT lisensi). audit iku dhewe
Enkripsi sumber-buka
Pustaka enkripsi kita iku sumber-bukak kanthi lisensi MIT. Aja percaya marang kita — verifikasi kode. Ora perlu percaya, mung matematika.
Situs resmi ing GitHub | Lihat sumberSiap kanggo transkripsi kanthi aman?
Upload file pisanan gratis. Transkripsi pribadi kasedhiya ing Pro lan Business rencana.
Ngawiwiti transkripsiPitakon kang asring diajukake
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.