Rescales a time column within each segment of a landmark set so multisyllabic
contours share a syllable-aware time axis. Adds two columns, named from the
set:
Usage
normalise_time_landmarks(df, time, set)
Arguments
- df
Data frame with the time column and the set's _start / _end
(and optionally _i) columns, e.g. from attach_landmarks().
- time
Name of the raw time column (seconds, matching the landmark units).
- set
Landmark-set base name (e.g. "syllable").
Value
df with the two columns appended. Returned unchanged when the
required columns are absent.
Details
<set>_t01 — time rescaled to 0-1 within each segment (segments overlap;
good for comparing segment shapes against each other).
<set>_tseq — sequential time, (<set>_i - 1) + <set>_t01, so segments
lie end to end across the word (a word-level time axis that keeps order).
