Speaker recording support

[KalobTaulien]

The World Conference floor voted in favour of creating a new feature in the app that supports ~5 speaker recordings. This would fulfill a referral from the Trustees.

These are my initial thoughts so far, which is all up for debate and open for discussion (as always!) and in no particular order.

• This would be a new application in for the API codebase with an API endpoint (Backend using Rest Framework)
• We only ever accept and stream recordings IF there is a legal waiver signed and ready to be shown at a moments notice.
• The app will require routing from the API data.
• A streaming service such as SoundCloud or something big enough to test out that embeds easily and without ads. This would be a test run, and if it's not an acceptable fit we can migrate 5 recordings very easily (Dependency).
• Release the current version of the app before adding more to the main branch. That will help keep this project tidy.

Here are some surface level requirements and dependencies:

Backend

• [ ] Possibly need an API key for a streaming service.
• [ ] Audio needs to be streamable (usually storing an object ID of some kind)
• [ ] Need a place to indefinitely save the waiver.
• [ ] Might as well add an approved BooleanField while we're at it. (One less migration in the future)
• [ ] API output using Django Rest Framework and serializers.

Frontend

• [ ] Need react native routing after fetching recording locations from the API
• [ ] Embeddability to stream audio from the service that's chosen/tested.
• [ ] We need a disclaimer on the RecordingListingScreen that says the speakers do not represent C.A. and their opinions are strictly theirs.

Legal

• [ ] Need copies of waivers that are within the hands of the WSOB; link stored in the database.

Dependencies

• [ ] Which recordings are we going to use and who selects them? (Not me)
• [ ] Choose an audio streaming service that can be embedded into the C.A. App. (No reinventing a simple audio player)

Kalob to look into some of this more and report back in this Issue for transparency.

[KalobTaulien]

SoundCloud

Feasibility: Low

• We get 3 hours on the free tier, and then it's CAD $10.42/month to get unlimited.
• Embedding using standard protocols is difficult.
• SoundCloud seems to have closed down the creation of new apps in order to work on their platform (from their blog and other news)

I will continue to look into SoundCloud as they would have been the best option, since this is their specialization. Any new information I will report back here.

[KalobTaulien]

Podbean

Feasibility: High The original referral looked at Podbean as an option, but the company seems quite small at this time. However, at a second glance it looks very viable for streaming. I will admit I was wrong to judge Podbean so quickly.

• We get 5 hours total on the free tier. That's enough for us to test out Podbeam with ~5 speaker tapes.
• $9/month afterwards and unlimited "downloads" (playbacks). This is [probably] the best pricing out there I've seen so far.
• Has an API that's relatively easy to use. Just for fun I was able to create a Python wrapper for it.
• The API returns a full URL to a raw audio file, such as username.podbeam.com/thing/speaker.mp3 -- this is great news for the app.
• They seem to have decent account security as well.
• We can add titles and richtext to each speaker via Podbean's interface, meaning the WSOB or whoever ends up managing this requires less developer support.
• No ads.
• Stats dashboard (when upgrading). Stats are always fun.