Open datasets — Perit

5.1 hFree on Hugging Face

15.5 hFull set, on request

48 kHz Stereo, 16-bit PCM

CC-BY-4.0 Licence on the open set

What is in it

Four things that make it usable for turn-taking.

These are properties of the recordings, measured per file and published with the set. No claim here is about how a model trained on it performs.

  1. One speaker per channel
    Every conversation is a true stereo pair, each speaker recorded on their own channel rather than split apart afterwards. Per-channel voice activity is ground truth, so turn boundaries do not have to be diarised first. Which channel holds which role varies by file, so the measured speech onset for both channels ships with every row.

  2. Full-duplex, with the overlaps left in
    Barge-ins, back-channels and interruptions are kept as recorded. Nothing is gated, ducked or edited to tidy the turn boundaries, because those boundaries are the signal.

  3. 48 kHz throughout
    Resampled to a single rate across the whole release, and loudness-normalised to EBU R128 so files from different sessions can be mixed in one training batch.

  4. Native US English, scripted scenarios
    Customer-support role-plays across healthcare, billing, airline, insurance and IT support, performed by four recurring speaker pairs (eight performers). Agent and customer roles swap between files.

Request the full set

The other 10.4 hours.

Tell us who you are and what you are building, and the link comes back straight away — no call, no approval queue. We ask because knowing who trains on this is how we choose what to record next.

The link is sent straight away. No sales sequence.

Consent and licence Cleared

Every speaker in this set was engaged and paid by us, and consented in writing to the recording being published and trained on. Nothing here is scraped, and no end customer is recorded: these are performed role-plays, so there is no personal data to redact.

How we handle consent

Need a locale that is not here

This set is native US English. We record the same duplex format in other locales to order, including Indic languages, with the channel separation and consent terms unchanged. Say which one on the form and we will tell you what exists and what it would take.

Commission a collection

If you train on it, say so.

The open set is CC-BY-4.0: use it commercially, fine-tune on it, redistribute it, as long as Perit is credited. If you publish a model or a paper that used it, tell us and we will point at your work from here.

perit-ai/perit-duplex-support-en