Security & Privacy
Nezve izvo zvinoitika kune yako audio uye transcripts pazuva nezuva. Hapana kushambadza fluff - chete sezvainoshanda.
Maitiro matatu ekuchengetedza ruzivo rwemunhu
Chimiro
Nese mushandisi, nese chirongwa — kusanganisira chisina kubhadhara
- HTTPS (TLS 1.3) yese data munzira
- Audio yakagadziriswa mushure mekugadziriswa
- Transcripts akachengetwa mu database yedu
- Tinogona kuverenga zvakachengetwa transcripts
- Data kamwe chete yakatengeswa kana kushandiswa kwedzidzo
- Dzvanya pane yako data nguva dzose
Transcript
Pro+ Available on Pro uye Business zvirongwa
- Zvese zviri muStandard, pamwe ne:
- Transcript encrypted in your browser (AES-256-GCM) before saving
- Tinochengeta chete data rakavharwa — hatigone kuiverenga
- Key inobva pa password yako, haizoendeswa kwatiri
- ⚠ Audio ichiri kugadziriswa panzvimbo yedu ye servers panguva yekushandura
Private Cloud / Self-inochengetwa
Full isolation — from $99/mo
- Audio haizombosiya yako infrastructure
- Kushandura kunoitwa neGPU yako
- Hapana data rakaendeswa ku STT.ai servers
- Air-gapped rutsigiro rwakakodzera
- True end-to-end privacy
Chii chinoitika kune yako data
A yakajeka, step-by-step kuparadzwa kwenzira yako audio uye transcript zvinodzorwa.
Standard (vashandisi vese)
1
Iwe uno upload audio kana kurecord live
Faira rako rinoendeswa kuburikidza ne HTTPS (TLS 1.3) kune yedu GPU server yekushandurwa.
2
Audio inoshandiswa mupfungwa
AI yedu inoshandura mavhidhiyo ako paGPU. Mavhidhiyo anochengetwa mupfungwa panguva yekuongorora — haasi kunyorwa padhiski — uye anobvisa kubva mupfungwa mushure mekuongororwa.
3
Transcript inochengetwa mudatabase yedu
The text transcript, timestamps, uye munyori mapepa akachengetwa kuti iwe unogona kuwana navo gare gare. Tinogona kuverenga iyi data (ichi ndicho nzira kutsvaka, AI zvinyorwa, uye kugovera basa).
4
Unogona kudzima zvese chero nguva
Kubvisa transcripts siyana kana yako account yose kubva Privacy Settings. Kubvisa ndechimwe uye zvinoitika zvisati zvaitika. Midziyo yePrivacy
With Private Transcript enabled
Maitiro 1-2 akafanana — audio yako inofanira kugadziriswa panzvimbo dzedu dzekuchengetera kuti iite transcript. Chimwe chinhu chinosiyana ndechekuti:
3
Transcript inochengetwa mubrowser yako pasina kumbobvira yachengetwa
Pashure pekushandurwa, chikonzero chinodzokera kubhurawuza rako. Bhurawuza rako chinoichengeta muAES- 256- GCM nekushandisa chiratidzo chakabva papassword yako (PBKDF2, 100K iterations). Chiratidzo chakachengetwa mu blob chinoendeswa kumaseva edu kuti chichengetwe. Hatifaniri kuona kana kuchengeta chiratidzo chekuchengetwa.
4
Tichachengeta chete encrypted data
Database yedu ine chete blob yakavharwa. Hatigone kuibvisa. Kana database yedu yakanga yavharwa, matranscripts ako aizove asina kunyatsoverenga.
Zvakanakira: Private transcript inodzivirira yakachengetwa transcript. Panguva ye transcription maitiro, yako audio inogadziriswa panzvimbo dzedu dzeservers kuti iite mashoko. Kana chikonzero chako chekukuvadza chichida kuti audio irege kusangana nenzvimbo dzechitatu, funga nezve Private Cloud kana Self-Hosted. Private Cloud
Chii Tinogona uye Chii Hachigone Kuona
Tinogona kuona (ne Private Transcript)
- Yako yakachengetwa transcript tenzi
- Mazita evanhu vanotaura kana mazita ezvinyorwa (akachengetwa)
- Timestamps kana data reword-level (yakachengetwa)
- Yako encryption key kana password
Tinogona kuona (anenge nePrivate Transcript)
- Audio yako panguva yekushanda (yakavharwa mushure me)
- Faira zita, saizi, nguva (metadata)
- Chirungu chakawanikwa, chigadzirwa chakashandiswa
- Kunyora
- Kuwana ruzivo nezve account yako
Technical Details
| Chivharo algorithm | AES-256-GCM (yakavimbika encryption) |
| Kubva kwechiratidzo | PBKDF2 ne SHA-256, 100,000 iterations |
| IV (nonce) | Random 12 bytes per encryption (never reused) |
| Kuchengeta kwechiratidzo | Haasi kuchengetwa — inogadzirwa kubva pa password pa ese ma sessions |
| Kufambisa encryption | TLS 1. 3 (HTTPS) + HSTS (1 gore, preload) |
| Audio retention | Yakagadziriswa mupfungwa, haizonyorwa pa diski, ichagadziriswazve |
| Kuvandudza | Web Crypto API (browser-native, hapana kunze malibrary) |
| Chinyorwa chemutauro | github.com/sttaigit/stt-encryption (MIT License) |
Private Transcript Trade-offs
Private transcript i opt-in nekuti encrypting yakachengetwa transcript inorambidza zvimwe zvinhu:
Inoshanda ne encryption
- Kuona yako transcripts
- Kutumira kunze (TXT, SRT, VTT, nezvimwewo)
- Kurodha pasi
- Kuchinja (yakavharwa mubrowser)
Hapana ne encryption
- Server-side search across transcripts
- AI zvinyorwa uye kufona (server haigone kuverenga data)
- Kugovana nevanhu kuburikidza nelink
- Kushanda pamwe nechikwata
Unofanira Audio Kuti Hapana Kudzokera Your Servers?
Transcript yakachengeteka inodzivirira transcript yakati yarara, asi mashoko anoenderera mberi achienda kuburikidza neGPU yedu panguva yekushanda. Kana zvinodiwa zvekuchengetedza kana zvekutevedzera zvako zvichida kuti mashoko asingazombosvibiswe nenzvimbo dzechitatu, heano sarudzo dzako:
Private Cloud
$499/mo
Dedicated GPU server inotungamirirwa nedu. Audio yako haizombosiya yako yakazvimirira nharaunda.
- Yakazvimirira A100 GPU
- Isolation — hapana kugoverwa kweinfrastructure
- Audio yakagadziriswa chete pahardware yako
- Full API kuwanikwa + SLA
Yakachengetwa nemunhu
$99/mo
Docker image. Your servers. Your GPU. Nothing leaves your network.
- Docker — inofamba pa chero NVIDIA GPU
- Air-gapped rutsigiro - hapana internet inodiwa
- Model updates zvakabatanidzwa
- Full control, full privacy
Isu Kuvimbisa (Vashandisi Vose, All Plans)
- Audio mafaera anogadziriswa mukati me24 mazuva ne default. Kugadziriswa muGPU memory, zvinongedzo zvemutauro zvinotorwa mushure mekunyorwa. Chimwe chete chekuchinja: kana iwe ukasarudza "Contribute corrections + audio to Voice Lab" muPrivacy Settings, zvinongedzo zvinochengetwa kwemazuva makumi manomwe nemana, uye zvinongedzo zvinodzoserwa neCron. Midziyo yePrivacy
- Data rako harizoshandiswa kweAI kudzidziswa kana iwe uchiti unozviita muPrivacy Settings (kudzidziswa kwekushandura mazita chete uye/kana kuongororwa kwemashoko muVoice Lab — zvese izvi zvinovharwa ne default). Midziyo yePrivacy
- Hatina kutengesa data rako. Kana chero munhu.
- All traffic encrypted in transit kuburikidza ne TLS 1.3 ne HSTS.
- Dzvanya pane yako data nguva dzose kubva kuPrivacy Settings kana nekudzima yako account. Midziyo yePrivacy
- Chinyorwa chekunyora chinovhurika-chigadzirwa — ongorora iwe pachako (MIT license). ongorora iwe pachako
Open-source encryption
Nhasi, cryptography library yedu inowanikwa pasi peMIT license. Usatidzai, edzai kuongorora kodhi. Hapana kutenda kunodarika, chete dzidzo.
View on GitHub | Ona _chizvarwaReady to transcribe safely?
Upload yako yekutanga faira yemahara. Private transcripts anowanikwa paPro uye Business mapurogiramu.
_Kutanga KunyoraZvimwe zvinobvunzwa kakawanda
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.