Audacity make it possible to annotate sound files with labels that can be exported as a .tsv file with .txt extension. This function convert result to dataframe.
audacity_to_df(file_name)
file_name | file_name string with a filename or path to the .txt file produced by Audacity |
---|
a dataframe with columns: content
, time_start
,
time_end
, source
.
#> time_start time_end content source #> 1 0.2319977 0.3953891 sssw test_audacity.txt