Flying fox recognizer #239
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Branch for Issue-#238
Issue-#238 Further work on the flying fox recogniser. Refactored a method used for frog recognisers and made it generally available in the class StandardSpectrograms.cs
Issue-#238 Main work was to add a new method to AcousticEvents.cs in order to find events located at a score maximum. All such events have the same (user defined) time duration centred on the maximum.
REmoved a lot of unnecessary commented code. Also improved call detection. Next step is to add additional contraints to the call spectrum to match that of bat call.
|
@towsey is this ready for review? |
Issue #238: change method which makes first attempt to look for FF chirps. Start second method which filters those on the basis of the similarity of their spectral profile to that of a flying fox.
Issue #238 This recogniser now works well on a ten minute recording containing 14-16 FF calls. TODO: The search parameters need to be read from config file. At moment hard coded.
Issue #238 Add Wingbeat profile to recogniser. It picks up wingbeats but also some human speech. Also set up to read parameters from the config file.
Did not work so far.
Issue #238 Do final work on Flying Fox wing beat detection. Produce event spectra for experimental debug purposes.
Issue #238 Reviewed FF recogniser on a four extra recordings. In particular worked on a 60 minute recording and was able to pick up most ac events of interest with few FPs and FNs.
atruskie
commented
Aug 27, 2019
There was a problem hiding this comment.
@towsey, lots of goof work here.
Most of the changes requested are just nits and cleanup.
However, the largest omission is a lack of unit tests.
Test are expected for the following:
- The entire pulse train class (using no audio)
- The
SpectrogramTools.GetSonogramPlusChartsmethod (using no audio) - And the
AcousticEvent.GetEventsAroundMaximiamethod (again, no audio required)
Additionally, there is no mention of the resource packet that was used to train/test this recognizer. Can you please describe where we can find a zip file with short segments of audio (no more than a minute) along with associated true/false labels for each expected event generated.
Thanks
src/AnalysisConfigFiles/RecognizerConfigFiles/Towsey.PteropusSpecies.yml
Show resolved
Hide resolved
src/AnalysisConfigFiles/RecognizerConfigFiles/Towsey.PteropusSpecies.yml
Outdated
Show resolved
Hide resolved
Issue #238 work on changes suggested by Anthony.
|
What has happened. I am working on all the issues you raised. Committed some yesterday - still more to go! |
|
Sorry - accidentally deleted too early. I'm on a mistake-making-productivity-steam-train! I've restored the branch. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Closes #238