Speaker diarization + transcription

Who said what.
Transcribed.

Pinna detects who-spoke-when and turns your call and meeting audio into speaker-labeled, time-aligned transcripts. Submit by web, email, or API — and get back a transcript that tells the voices apart.

Get started → See how it works
עברית Hebrew العربية Arabic English Русский Russian
Capabilities
Diarization and transcription, in one pass
Two steps run together: diarization labels who is speaking, transcription turns each turn into text. You get back one transcript — segmented by speaker, aligned to the recording's timeline.

Speaker diarization

Detects how many speakers are in the audio and labels who-spoke-when automatically — no manual tagging, no guessing.

Whisper-class accuracy

Whisper-class accuracy in Hebrew, Arabic, Russian and English — including the languages cloud tools get wrong. Speaker diarization built in.

Four languages

Hebrew, Arabic, English, and Russian — including right-to-left scripts. Auto-detected, or force a language when you know it.

Three ways in

Upload in the web app, email an attachment, or call the API. Same engine, same speaker-labeled result, whichever you use.

Offline-capable engine

The core runs with no third-party cloud-API dependency — the foundation that makes on-prem and airgap deployment possible (available with Enterprise).

Time-aligned output

Every speaker turn carries a timestamp, so you can jump straight to the moment in the recording a line came from.

How it works
Audio in, speaker-labeled transcript out
Submit
Web upload, email, or API
Pinna processes
Diarize + transcribe
Get the transcript
Speaker-labeled, time-aligned
transcript.txt
00:00:04Speaker 1Thanks for joining — can everyone hear me okay?
00:00:07Speaker 2Loud and clear. Let's get started.
00:00:11Speaker 1Great. First item is the rollout timeline…
Submit audio
Three ways to send us audio
Pick whatever fits how you already work. Every path runs the same engine and returns the same speaker-segmented transcript.
1

Web upload

Drag a file into the app, watch the job progress, and download the transcript when it's done. The simplest way to get started.

In the personal area at account.pinna.im.
2

Email

Send the audio as an attachment and get the transcript back — no app needed. Perfect for forwarding a recording the moment a call ends.

Mail it to <you>@in.pinna.im. The sending address has to be on your account allowlist.
3

API

Submit jobs programmatically and pull results into your own systems. Built API-first, so automation is a first-class path, not an afterthought.

Authenticated with an API key you manage in the personal area.
Pricing
Pay for hours, not features
Every plan includes all four languages and speaker diarization. Plans differ on audio hours per month and delivery mode.
Free
$0 /mo
1 hour per month. All languages, diarization, web cabinet. No card required.
Get started
Starter
$12 /mo
10 hours included, then hourly. Full export.
Get started
Enterprise
Custom
On-prem / airgap, SSO, SLA, compliance docs.
Contact us

Every plan includes all languages and diarization. You pay for hours and delivery mode, not features.

Who it's for
Built for serious audio — in the languages the cloud gets wrong.
For lawyers, clinicians, investigators, journalists, newsrooms and regulated call centers — wherever recordings are privileged, sensitive, or compliance-bound. Process in our secure cloud, or with Enterprise run fully on-prem / airgap, so the audio never leaves your perimeter.

No cloud-API dependency

The core diarization and transcription run on a self-contained engine. Your audio doesn't get handed to a third-party transcription API to do the work.

On-prem & airgap · Enterprise

Because the engine is offline-capable, Enterprise can run Pinna entirely on infrastructure you own — including airgapped networks — so audio never leaves your perimeter.

You control access

Email ingress is allowlisted per account, and the authenticated app manages API keys and job access — so only the people you authorize can submit and read.

FAQ
Questions, answered
The things people ask before sending us their first recording.
Does Pinna identify who is speaking?
Yes. Speaker diarization detects how many speakers are in the audio and labels who-spoke-when automatically — no manual tagging. Every speaker turn is time-aligned, so you can jump to the moment in the recording a line came from.
Which languages does Pinna transcribe?
Hebrew, Arabic, English, and Russian — including right-to-left scripts — with Whisper-class accuracy, including the languages cloud tools tend to get wrong. Language is auto-detected, or you can force one when you know it.
Is my audio sent to a third-party transcription API?
No. The core diarization and transcription run on a self-contained engine with no third-party cloud-API dependency — your audio is not handed to an outside transcription service to do the work.
Can Pinna run on-premises or in an airgapped network?
Yes, with the Enterprise plan. Because the engine is offline-capable, Pinna can run entirely on infrastructure you own — including airgapped networks — so the audio never leaves your perimeter.
How do I send audio to Pinna?
Three ways, all running the same engine: upload in the web app, email the audio as an attachment to your allowlisted <you>@in.pinna.im address, or submit programmatically with an API key. You get back the same speaker-labeled transcript whichever you use.
Who can access my audio and transcripts?
Only the people you authorize. Email ingress is allowlisted per account, and the authenticated app manages API keys and job access, so only authorized senders can submit and only authorized users can read results.
Get started
Turn your next recording into a transcript
Create an account, upload your first file, and see who-said-what in minutes. The free plan needs no card.
The Pinna app — upload, jobs, transcripts, API keys, and the email allowlist — lives at account.pinna.im.