asearfos / chipper Goto Github PK
View Code? Open in Web Editor NEWSoftware (with Kivy GUI) for semi-automated segmentation and analysis of birdsongs or other acoustic signals.
License: GNU General Public License v3.0
Software (with Kivy GUI) for semi-automated segmentation and analysis of birdsongs or other acoustic signals.
License: GNU General Public License v3.0
This problem come from how we are defining a syllable. Since a syllable is any signal between an onset and offset (with no frequency bounds) then any noise above or below what the eye would call the syllable is being counted when max and min are being measured.
Only idea to improve this as of right now is to use the note threshold and remove anything below it before conducting syllable calculations (including similarity check?).
pad really short songs to be more "regular" looking in the window rather than very stretched
Choose consistent color for noise among all songs in syllable similarity widget (dark gray maybe).
If easy, could color within syllable noise white like in the note threshold widget and color between syllable signal as dark gray.
make parameter output in .txt file in Hz and ms for user readability
Analysis could error out without providing any feedback to user. Need to put a try : catch block and let user know using the spinner.
in syllable segmentation, fix crash when submit is hit but there are no syllables (no lines)
Change the language so it is clear this is more about figuring out what is considered noise within syllables so that it is not considered in calculations.
only if this is an easy change
If a single file is selected instead of a folder the program crashes.
temporary save upon crash or exit
We added a catch for this in the actual analysis but Chipper also crashes if this happens in the syllable similarity threshold widget if the note size threshold is set too large (no notes in syllable).
if reset parameters, they currently go to the ones the user input; if parameters were found from a previously chippered gzip, they should go back to those instead (not sure how to do this without adding params as a self variable)
add output of frequencies and durations, not just average measurements
Rough frequency measures (rounded) were added at 0Hz and at half the full frequency of the spectrogram.
constrain where to look for gzips if using previous parameters, or give user option of which gzips
Instead of one output analysis file, have two with one being all note measurements. This can be labeled so as to warn the user to be careful/critical of these measurements (especially for noisy files), be sure they use the widgets, and check the outputs for reasonable values.
right now only computes as if the WAV files are 44100 Hz even if they are not
flag for too large of file; split really long songs, subsequently stitch back together before writing (@asearfos can integrate kivy popup to warn users not to change the sliders between songs)
general garbage collection as well as checking variable handling when done and returning to landing page from each widget (each radio button)
This happens in syllable segmentation when you have already added or deleted a syllable onset or offset and hit enter. If you hit enter again, it crashes.
Make a test to make sure that the values are the same (similarity between two defined syllables).
Make a test to make sure frequencies dont overlap.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.