Transcription thathears every word.
Real-time and batch on the same API. 50+ languages. Speaker diarization, punctuation, and word-level confidence by default.
Drop audio · or record
mp3 · wav · m4a · up to 2GB
Underlined = low confidence. Click to edit.
96.4% on clean en-US conversational audio.
Streaming output in under 300ms.
50+ languages with code-switching.
AI purpose-built for your industry’s toughest challenges.
Your customers’ problems don’t look like everyone else’s. Your AI shouldn’t either. OneInbox is tuned to the workflows, urgency, and stakes that define your industry.
Why teams switch.
Same audio, side-by-side. The same things keep showing up.
| OneInbox | Others | |
|---|---|---|
| Streaming and batch on one API | ||
| Diarization without extra config | ||
| Word-level confidence | ||
| Code-switching mid-utterance | ||
| Pricing per minute, not per request |
Trusted by teams that ship.
Better diarization than the incumbents we tested. We swapped our pipeline in a single day.
Aiko Watanabe
CTO · Lattice Health
Frequently asked.
mp3, wav, m4a, flac, ogg, and webm. Files up to 2GB. Streams over Opus or PCM.
We benchmark at 96.4% on clean en-US conversational audio and 91% on call-center audio with background noise. Word-level confidence lets you flag the rest.
Per minute of audio processed. The same rate for streaming and batch. Volume discounts apply at 10k minutes per month.
Only if you opt in. Default is zero-retention. Enterprise customers can pin storage to a region or run fully self-hosted.
Start transcribing in under a minute.
Free tier covers your first 10 hours. No credit card.