A complete transcription platform

🎯 Accuracy you can trust

Transibo runs faster-whisper (CTranslate2) with the model you choose, up to large-v3. Voice-activity detection trims silence, and a multilingual, code-switching aware mode detects when a recording mixes languages and transcribes each segment in the right one.

• Configurable beam size, best-of and patience for quality/latency trade-offs
• Per-segment language detection with an ambiguity view
• Confidence signals (avg logprob, no-speech probability) retained per segment

📦 Batch queue with auto-resume

Long media is processed on a Postgres-backed queue using FOR UPDATE SKIP LOCKED with heartbeats and a watchdog. If a worker crashes or is redeployed mid-job, processing resumes from the last completed segment — no lost work, no duplicate charges.

• Durable, restart-safe job processing
• Live progress and per-pass status
• Priority lanes so paid plans and short clips jump ahead

💬 Subtitles & SRT

Every job produces timestamped segments, a plain-text transcript and a ready-to-use .srt subtitle file. Download them from the dashboard or pull them straight from the API.

🌍 Translation

Translate transcripts and subtitles into other languages with built-in NLLB translation. Keep the original and the translated SRT side by side.

⚡ API access

Automate everything with the /v1 REST API. Create per-org API keys (scoped, hashed at rest), transcribe short clips synchronously, enqueue long jobs, and receive HMAC-signed webhooks on completion.

• POST /v1/transcribe — sync short-clip transcription
• POST /v1/jobs / GET /v1/jobs/{id} — async jobs
• OpenAPI schema at /docs for client generation

Developer docs →

👥 Team & org workspaces

Each account gets an organization with role-based access (Owner / Admin / Billing / Member), email invitations, per-org quotas and usage, retention controls, and a full audit log. Tenant isolation is enforced on every query — your data is only ever visible to your org.