Skip to contents

Calculate per-speaker statistics based on the .its file

get_lena_speaker_stats.Rd

Only analyzes conversation blocks (blkType == 'Conversation'). Uses point-in-time matching where segment start time must fall within interval boundaries (unlike add_lena_stats which uses overlap-based proportional allocation).

Usage

get_lena_speaker_stats(its_xml, intervals, time_type = c("wall", "wav"))

Arguments

its_xml

XML object created by rlena::read_its_file.

intervals

A tibble with columns depending on time_type:

  • For wall: interval_start, interval_end (POSIXct), interval_start_wav (ms)

  • For wav: interval_start_wav, interval_end_wav (both in ms since wav start) Can contain other columns.

time_type

Either wall or wav. If wall (default), expects columns interval_start, interval_end (POSIXct timestamps), and interval_start_wav (milliseconds since wav start). Will calculate interval_end_wav dynamically and drop it from output. If wav, expects interval_start_wav and interval_end_wav (both in milliseconds since wav start).

Value

a tibble with the following columns:

  • interval columns: same as in the intervals input tibble,

  • spkr: speaker identifier from LENA

  • adult_word_count: non-zero for FA* and MA* speakers only

  • utterance_count: for CH* - the sum of childUttCnt, for everyone else - the number of conversation segments

  • segment_duration: total duration of segments for this speaker in this interval