Add new metrics for each speaker

pierreaubert / spinorama

A library to display and compare spinorama (speakers measurements) graphs.

GNU General Public License v3.0

119 stars 13 forks source link

Add new metrics for each speaker #289

Open pierreaubert opened 3 months ago

pierreaubert commented 3 months ago

Is your feature request related to a problem? Please describe.

Request to add F12 and F9 on top of F6 and F3 #284
Request to add value for ERDI and SPDI #284
Request to add scores for Directivity #287

Describe the solution you'd like

Compute the various scores and add them to json files
Expose some of the scores into
- Large ranking tables
- Add values into the speaker data page

Additional context There are already some computed scores for DI but they are not exposed.

https://github.com/pierreaubert/spinorama/blob/ecf40662fda899e3cd8093b59b0bf09d748ee42c/src/spinorama/compute_misc.py#L298

pierreaubert commented 3 months ago

Made some progress, F6->F12 are computed (I forgot F15 but will fix that).

pierreaubert commented 2 months ago

See https://docs.google.com/spreadsheets/d/1dQe59JchQ33V0r1J9kjgnACyouN6Q_HkJHh56iaJHm4/edit?gid=1829242726#gid=1829242726 to view all values.

Odjur commented 2 months ago

That's awesome, thank you. Any chance to add DI to the spreadsheet?

pierreaubert commented 2 months ago

DI is coming. Question is what do you want? I have added the std dev of SPDI. Lets see if that got reasonable results first.

Odjur commented 2 months ago

The simplest would probably be what VituixCAD does: ERDI SM and Slope, and SPDI SM and Slope, where SM is "Smoothness (r^2) in amplitude response based on a linear regression line through 100 Hz - 16 kHz" and Slope is "Slope of best fit linear regression line through 100 Hz - 16 kHz"