Get paid for the podcast audio you already own.
We license conversational, studio-grade podcast audio to companies building speech and voice AI. If you’ve recorded a podcast, narrated an audiobook, or run a studio that produces interview content, we want to talk. Real contracts. Direct payouts. Your name stays on every release.
Conversational audio that sounds like real people talking.
The closer your audio is to two real people having a real conversation in a real room, the more valuable it is to us.
We’re actively buying
- Long-form interview podcasts
- Panel discussions and roundtables
- Solo monologue and narrative shows
- Audiobook narration
- Scripted dialogue and roleplay
- Animated character voice work
- Multilingual content in any language
- Regional accents and dialects
We pay a premium for
- 50+ hours from a single show or speaker
- Multi-language back catalogs
- Native speakers of under-represented languages
- Professional studio recordings
- Existing aligned transcripts
- Multi-track per-speaker WAV files
What we cannot accept
- Recordings of anyone without explicit consent
- Phone calls, support recordings, covert audio
- Audio you don’t fully own or control
- Music-heavy content where speech is secondary
- Audio with PII or confidential business info
What “studio-grade” actually means.
If your setup matches the “preferred” column, you’re in great shape. If it matches “minimum,” we can still work with you.
| Spec | Preferred | Minimum |
|---|---|---|
| Sample rate | 48 kHz | 44.1 kHz |
| Bit depth | 24-bit | 16-bit |
| Format | Multi-track WAV | WAV, FLAC, AIFF (lossless) |
| Duration | 5+ hours per submission | 20 minutes continuous |
| Microphone | Shure SM7B, Rode NT1, Sennheiser MKH 416 | Condenser or broadcast dynamic |
| Environment | Professional studio | Treated room, no echo |
| Background noise | Inaudible | Minimal |
| Overlapping speech | Multi-track per speaker | None during the same channel |
| Transcript | Word-level aligned, any format | Optional |
Don’t have aligned transcripts? Don’t worry — we run our own transcription and diarization in-house, and we’ll generate them for you. You don’t need to do extra work to qualify.
Three ways we pay creators.
Compensation depends on hours, language, accent, and how rare or in-demand that combination is. The market is moving fast, so we don’t publish a fixed rate card — but here’s the rough shape.
Catalog license
- Per-hour rate, paid up front
- You retain ownership
- One-time license fee for AI training use
- Most common deal type
Catalog + revenue share
- Paid up front + share of downstream revenue
- Best for large back-catalogs
- Ongoing income as we relicense
- Quarterly payouts
Exclusive license
- Buyer gets exclusive use of your audio
- Significantly higher rate
- Multi-stage payout (signing + delivery + milestones)
- For rare languages or distinctive voices
Direct deposit · Bank transfer · Wise · PayPal · Paid in USD, EUR, or your local currency
No surprises with your audio.
What we will do
- License the audio to vetted AI companies for training speech, voice, and conversational models (ASR, TTS, voice agents)
- Generate aligned transcripts and speaker metadata to make the audio more valuable
- Store the audio in encrypted, access-controlled cloud storage
- Keep a documented chain of consent linking every file back to a signed release
- Pay you on time
What we will never do
- Create a voice clone of you or anyone in your audio under a standard license
- Sell, share, or sublicense your data to a buyer who hasn’t signed our standard MSA
- Use your audio to train a model that impersonates you specifically
- Hand your data to a buyer without redacting things you’ve asked us to redact
- Lock you in — you can revoke consent at any time
You always remain the owner of your audio. We’re licensing it from you, not buying it. If something changes — you take a show down, a guest revokes, your business pivots — tell us, and we’ll update the consent record and notify any downstream buyers.
From submission to first payment.
Submit
Fill out the form below with a sample (a few minutes is fine), your languages, and roughly how many hours you have.
Vetting call
We listen to your sample and confirm fit, usually within 3 business days. 20-minute call.
Verification
We verify you’re the rights holder. For voice work, we use voice biometrics — a short script you record so we can confirm the speaker on the catalog matches the speaker on the contract.
Contract
We sign a standard creator agreement. Plain English, no surprises. Average turnaround: 3–5 business days.
Delivery
You upload your catalog to a private, encrypted bucket we set up for you. We handle transcription, diarization, and QA.
Payment
Funds hit your account on the schedule in your contract. For most catalog licenses, that’s net-15 from contract execution.
Most creators go from first form submission to first payment in 2 to 3 weeks.
Tell us what you have.
Takes about 3 minutes. We’ll get back to you within 3 business days.
Questions creators ask us.
Do I keep ownership of my audio?
Yes. You always retain ownership. We license the audio from you under contractually-defined terms — we don’t buy it, and you can keep distributing your podcast normally on every platform you already use.
Will you make a voice clone of me?
No. Our standard creator agreement explicitly prohibits using your audio to train voice cloning models or to synthesize a model that impersonates you. If a buyer wants those rights, that’s a separate, much more expensive contract that we only sign with your explicit, written consent.
How do you verify I’m the actual person in the recordings?
Voice biometrics. After we agree commercials, we’ll send you a short script to record on the same setup you use for your podcast. We compare that recording to the voice in your catalog to confirm identity. Takes about 5 minutes on your end.
Do I need to provide transcripts?
No. Aligned transcripts add value to your audio (and in some cases bump your rate), but they’re not required. We run transcription, diarization, and QA in-house.
How much will I get paid?
It depends on language, accent, hours, recording quality, and exclusivity. The market is moving quickly, so we don’t publish a fixed rate card — but a typical English-language conversational catalog of professional studio quality currently lands somewhere in the hundreds-to-low-thousands per hour range. Rare languages and accents pay significantly more. We’ll quote you specifically after the vetting call.
How and when do I get paid?
Direct deposit, Wise, or PayPal. For standard catalog licenses, you’re paid on net-15 terms from contract execution. Larger or custom deals are paid on milestones we agree in writing before you sign anything.
What if a guest on my podcast doesn’t want to be included?
Tell us. We’ll either exclude their episodes from the dataset or run a redaction pass that removes their speech turns. You’re responsible for confirming consent from anyone whose voice is in the audio you submit, and our submission form asks you to attest to that — but we have a process for handling changes after the fact.
Can I revoke my data later?
Yes. You can ask us to remove your audio from any future deliveries at any time. Already-delivered datasets can’t be retroactively pulled out of model weights that have already been trained, but the revocation is logged in our provenance system and applies to all downstream contracts going forward. This is the same right-to-revoke model we offer our enterprise buyers.
Who buys this data?
Companies building speech recognition (ASR), text-to-speech (TTS), conversational voice agents, and multilingual voice AI products. We only work with buyers who sign our standard MSA, which includes IP indemnification, prohibits resale, and requires them to honor revocation requests.
What languages and accents are most in demand right now?
Always changing. Right now: under-represented languages (Hindi, Arabic, Swahili, Tagalog, Vietnamese, Bengali), regional accents within widely-spoken languages (Scottish English, Quebec French, Egyptian Arabic, LATAM Spanish), and any high-quality conversational audio in languages with limited podcast infrastructure.
I’m a podcast network or studio with dozens of shows. Can we do this at scale?
Yes — that’s actually our preferred deal size. Submit through the form above and mention “network deal” in the notes field, and we’ll skip ahead to a partnership conversation. We have a separate process for catalog deals over 1,000 hours.
Where are you based?
Headquartered in the United States. We work with creators worldwide and pay in USD, EUR, or local currency.
Your back-catalog is sitting on a hard drive. Let’s put it to work.
Submit your catalog and we’ll be in touch within 3 business days with a rate range and next steps.