This space captures VQ/HMM classification exercises on labelled whale song unit data.
Complete exercises with some level of model tuning.
A general presentation about VQ/HMM and its application on the initial set of exercises described here can be found at https://github.com/ecoz2/ecoz2-doc/blob/master/exploration/vqhmm-whale.pdf.
-
exerc00: On 10 labelled files.
-
exerc01: On a 4.5 hour recording.
-
exerc02: As exerc01 but with some underlying adjustments in programs parameters and file organization.
-
exerc03: related with the "windy" interval in the 4.5 recording:
-
exerc04: As exerc01 but with number of refinements iterations set to 120
-
exerc05: As exerc01 but with order of prediction P = 20 according to rule of thumb P = 4 + fs / 1000, where the sampling frequency fs is 16kHz in our case.
-
exerc06: As exerc05c but merging classes "I" and "I2" into a "II" class.
See signal-inspection.