For podcasts recorded one mic per person
Read your episode instead of listening to it.
Soundmancer transcribes each person's track on your own computer, and you edit the episode by reading the conversation. Strike a line and the audio goes with it. Then export the edit to Audition, Reaper, Audacity, stems or a finished mix.
- 0:41:07MayaSo the part nobody tells you about is the gear list. It changes every year.
- 0:41:15TheoHang on, my dog is at the door again. Sorry. Give me one second.
- 0:41:22TheoOkay. Back. You were saying the gear list.
- 0:41:26MayaRight. What I tell people now is: buy one good mic, and then um, stop buying things for a year.
- 0:41:34JuneWhich nobody does.
- 0:41:36TheoNobody. I have four mics in a drawer.
Cuts land in the quiet
Reading is much faster than listening. Cuts land where the speaker is already quiet, with a short fade.
Your DAW gets the edit
Export a session for Audition, Reaper or Audacity for your polish pass, stems with one file per person, or a finished mix. All of them match each other and what you heard in the editor.
Nothing is uploaded
Transcription runs on your computer. There's no upload to wait for, and an unreleased episode never goes anywhere.
How an edit goes
Built for long shows with a lot of people on them: three or four hours, five to seven voices.
- Bring in the raw tracksOne file per person. Transcription starts straight away, and each track is its own speaker, so nothing has to guess who said what.
- Read and strikeThe conversation reads as a document. Strike what goes. Fillers are found for you, to cut one at a time or all at once.
- Tighten and clean upShorten the pauses across every track at once, set each person's level, and switch on per-track sound cleanup while it plays.
- ExportTo Audition, Reaper, Audacity, stems or a mix, with your markers carried into every one. Then master it: loudness, a true-peak limiter, tags, chapters and cover art.
When everyone talks at once
Crosstalk you can read, and cut one voice at a time
Most editors give you one script, and when people talk over each other they send you to a timeline. Cutting a word there cuts everyone who was talking at that moment. Soundmancer keeps every voice on its own track, so the transcript shows who talked over whom and where, and a strike takes out only that person.
- A quick "yeah" or "mm-hmm" sits in the sentence it was said under, the size of a word.
- Somebody cutting in shows at the word where they started. Point at it to see what they talked over.
- When people talk over each other for a few seconds, the transcript shows a column for each of them, and a click on a line strikes it.
- Strike one voice and the others keep talking. The edit tells you when a cut saves no time.
- Every overlap is sorted for you. There's nothing to tag.
- 0:12:04AnnSo we never actually agreed Bobmm-hmm on a number, and then in MarchCy Wait, we did agree.the whole thing fell apart.
- 0:12:15BobNo, forty was the draft, not the
- AnnIt was never signed
- CyI have the email!
- 0:12:19AnnOkay, okay. Let's look at the email.
The beta
Come test it with me
I'm opening the beta to a small first group. You get a free key and the app, and you can write to me directly. I ask for a problem report from the app whenever something goes wrong.
- Free during the beta.
- Unlimited transcription.
- Works offline once it's activated.
Unlimited transcription
Transcribe every hour you record. Other editors sell transcription by the hour or the credit; here it runs on your own computer, so there's nothing to count. You decide every cut, and every cut can be undone.
Trim and Master work on plain files
Trim removes the stretches where everybody is silent, from every track at once, so nothing drifts out of sync. Master sets the loudness of a finished mix and writes a tagged MP3 and an archive WAV.