Does anyone here have experience with analyzing data that has been recorded at different time periods, thus have different recording qualities? My question is whether the differing qualities might have an affect on how luscinia picks up signals, leaving results that may be false? If so, are there settings or a way to get around this?
For example, I am working with songbird spectrograms recorded in 1982, and comparing them to spectrograms recorded in 2012. Different recording equipment was used, thus audio quality is a little different. Signals tend to be a little more fuzzy in 1982 recordings, and clearer (maybe higher amplitude as well) in 2012 recordings. Even with a general noise removal and cleanup, I fear that there is enough difference where, when the analysis is run, syllables could be labeled as more different than they actually are.
Hello!
Does anyone here have experience with analyzing data that has been recorded at different time periods, thus have different recording qualities? My question is whether the differing qualities might have an affect on how luscinia picks up signals, leaving results that may be false? If so, are there settings or a way to get around this?
For example, I am working with songbird spectrograms recorded in 1982, and comparing them to spectrograms recorded in 2012. Different recording equipment was used, thus audio quality is a little different. Signals tend to be a little more fuzzy in 1982 recordings, and clearer (maybe higher amplitude as well) in 2012 recordings. Even with a general noise removal and cleanup, I fear that there is enough difference where, when the analysis is run, syllables could be labeled as more different than they actually are.
(@rflachlan)