Ask HN: Has anyone tried to do strava for music practice? | Hacker News Reader

Ask HN: Has anyone tried to do strava for music practice?

2 pointsflave4 comments

I've recently started playing the violin and guitar again after ten years of ignoring music (because I was forced to play as a child). During that time I started two technology businesses, one successful and one not, and I've learnt an okay amount about the 'tech start up' space.

When I practice, I have no record of what I practice, when and for how long - unless I write it out into a spreadsheet or airtable (which is what I've been doing). Time consuming, not standardised and takes more effort the more I practice. Hard to record that I practiced from bar x to bar y. Difficult to manually compute whether I've got good practice coverage over a whole piece.

Does anyone know of anything that works like Strava, can passively 'listen' to my practice and produce a sense of what I played and for how long. Produce 'snippets' of sound from each day so I can compare day to day (I do this manually right now).

I'm trying to work out whether this is something I could pursue as a start up (and scratch this bloody itch). Thanks!

I haven't run into this, but considered a variant on it by connecting Yousician to Beeminder -- self-binding for piano practice (https://blog.beeminder.com/akrasia/). I encountered a couple things that made it not worthwhile for me:

- I enjoy practicing, so the benefit from self-binding is reduced.

- Yousician doesn't have a supported API, just unofficial libraries written by users, so my connection would be brittle.

- I could probably only have tracked minutes per day or week, and while that's not bad, it's not a great measure of actual progress (and over half of my practice minutes are not using Yousician).

If Yousician had an API and Beeminder had native support for it, I'd probably have linked them and made a commitment. Somewhere in here, there's something I'd use :-) I think the best element of what you described is that it's completely passive yet works for anything. If I have to do a lot of session setup (like telling it which bars I'm playing), or run it on a computer instead of a phone/tablet, I don't think the benefit would be worth the effort.

For me, I'd accept whatever tradeoffs were required in order to make it completely passive (other than starting and stopping the app for a practice session). Even if that meant I only got heuristics and had to trust its inferences, that beats taking my hands off the keyboard to mess with it. For example, maybe it estimates my tempo accuracy based on how much my recording changes when quantized, and infers that more than X seconds of silence is the start of a new analysis period (new playthrough).

Feel free to contact me if you want a sounding board.

I'd love to chat - shall I reach on twitter? or send me an email (willnash at me dot com)

Tonara has some aspects of this, but not in the way you're imagining it I think. Would this be something that would be directed more at the professional (like symphony) musician market?

I mean, I see it being aimed at people like myself to start with (just because I kind of know I have this problem). So that would be people just above the beginner stage but below the professional stage. Hobbyist/amateur.

I have a feeling that student musicians would get a lot of value out of it as there's quite a big step change in how you should practice. i.e. at a student/pro level you learn to practice with a lot more structure and intention.

Thanks for that lead on Tonara! Which aspects where you thinking of?

Reply on news.ycombinator.com