0.1.1a1

• Initial version after excising from hvc
  (https://github.com/NickleDave/hybrid-vocal-classifier/commits/0c50144d75e3a3205db82add8b48302edbbed511/hvc/koumura.py)

changed

• The library now limits itself to returning annotation in a data structure
  similar to what's used in the BirdsongRecognition code.
  • e.g. parse_xml no longer returns an "annotation dictionary"
  • instead, the conbirt package
    handles translating the data structures to the "annotation dictionary"
    structure that it uses internally
• Convert tests to Python unittest format (instead of using PyTest library)

added

• Write README.md with usage
• to help work with conbirt package, the parse_xml function
  now has parameters to allow returning the paths to the .wav files as
  absolute paths, but by default it returns just the file names, as
  specified in the Annotation.xml files.