Map out user journeys and requirements for an online recording studio

[lpmi-13]

The idea is to have a repository that somebody can clone and run terminal commands to spin up an entire ready-to-record virtual studio.

It needs the ability to record audio, edit audio, create and update metadata for the audio (optional formatting for a particular podcast medium), and push to a particular podcast stream if necessary.

MVP will target spotify, cause that's what we're already doing, so we'll just set it up so it works for ourselves.

Open question as to whether we want to include automated account setup (probably not).

But definitely, once everything is set up, we should be able to enter credentials as environment variables, fire one command and be able to start recording.

...actually, maybe that's infeasible, so let's just start by listing the different pieces we need to fit together.

• zencaster for recording
• main video medium (we've been using zoom, but presumably zencaster can sync with other stuff)
• editing software (can zencaster do this?)
• editor for podcast feed XML
• a place to point the podcast feed at (we're using a github repo)

[benfoxall]

Oooooooh, yyyeeaaahhhhh!

The part I was interested in was replacing zencaster + some editing, and metadata (title description etc).

Zencaster only seems to have "automagical producing", which is cool for levels, but didn't fix the echo feedback thing (which was easy when you're manually doing it). The audacity app is amazing for manual stuff, though possibly too full-blown and powerful, which makes it easy to make mistakes. I want a limited set of options (silence a section of a single track, and cut a section of all tracks). By assuming that the tracks are aligned, you can cut out a bunch of things.

I'd been imagining you login (anon & securely with webauthn), invite others (no login needed), all media records locally then pushed up to S3. Then you can edit the audio and write metadata.

The metadata would use y.js crdt, so everyone would collaboratively edit, and I was really interested if you could do the same thing with editing the sound (so, different people can be editing different places simultaneously).

I hadn't thought about the part about distributing it at all! It'd be badass to output XML somehow. Pushing it up automatic would be ammmmazzzinnnnggg.

And oh, woah. Yes! Three js for immersive podcasting. Oh my. Oh my oh my.