One-off transcription tools are built for a single upload. templobe is built for the backlog: add up to 50 English audio files to a batch, and a speech model on your own Mac works through the queue while you do something else. Nothing is uploaded and nothing is metered by the minute.
50 files per batch. 2 GB per file, 5 GB per batch. mp3, wav, m4a. English only in this beta.
| What matters | templobe | Typical cloud workflow |
|---|---|---|
| Getting files in | Add up to 50 files to one batch | Upload, wait, repeat for each file |
| Throughput limit | Your Mac's CPU and memory | Plan limits, queue position, upload speed |
| Cost of a big backlog | Free for beta users | Scales with total minutes |
| Where audio goes | Stays on your Mac | Uploaded to a server |
| Works offline | Yes, after one sign-in | No |
| Failed file in the middle | Shows failed; the rest of the batch continues | Re-upload and restart the job |
| Export | txt, md, srt, vtt per file | Varies by provider |
| Data retention | You delete the local file; no copy elsewhere | Depends on provider policy |
Cloud services differ widely; this compares the common model, not one specific vendor. If you handle confidential recordings, check the provider's retention policy before uploading.
Select mp3, wav or m4a files. Up to 50 per batch, 2 GB each, 5 GB total. The files stay where they are — templobe reads them locally.
Each file is transcribed by a model installed on your Mac. Progress is per file, so one bad file does not take the batch down with it.
Open any finished file against its playback, then export as txt, md, srt or vtt. Start the next batch whenever you need.
Up to 50 files per batch, with a 2 GB limit per file and 5 GB in total per batch. When a batch finishes you can start another — there is no daily cap during the beta.
No. Batch work is included in the free beta. There are no per-minute credits and no higher tier you have to buy before a queue will run.
That file is marked failed and the rest of the batch continues. You can review the error per file instead of losing the whole job.
No. Files are read from disk and transcribed by a local model. There is no upload step, and the app works offline after the first sign-in.
It depends on file length and your Mac. In steady state the app processes roughly 40 seconds of audio per second of compute on the reference Mac, so a batch of short clips finishes quickly, while hours of long recordings take proportionally longer.
mp3, wav and m4a. Video files are not supported in this beta, and neither are languages other than English.
You sign in once to activate the app and download the model. That account never receives your audio, and it is not required to stay online while a batch runs.
It is a beta: English only, no speaker labels, no direct video import, and throughput is bounded by your Mac rather than a server farm. In exchange, batch transcription is free and private.
Beta users get free transcription for good. Leave your email to reserve the beta — no payment, no card. We'll send the signed build to the same inbox once it's ready. No newsletter, no drip sequence, no sharing your address.
We sent a confirmation link. Click it to confirm your address — your reservation is complete, and the beta build goes to the same inbox the moment it's signed.