enrich_rttm {voice}

Enrich RTTM files

Description

Enrich Rich Transcription Time Marked (RTTM) files obtained from 'voice::read_rttm'.

Usage

enrich_rttm(listRttm, silence.gap = 0.5, as.tibble = TRUE)

Arguments

listRttm

A list containing RTTM files.

silence.gap

The silence gap (in seconds) between adjacent words in a keyword. Rows with tdur <= silence.gap are removed. (default: 0.5)

as.tibble

Logical. Should it return a tibble?

Value

A list containing either data frames or tibbles obtained from standard RTTM files. See 'voice::read_rttm'.

References

https://www.nist.gov/system/files/documents/itl/iad/mig/KWS15-evalplan-v05.pdf

See Also

voice::read_rttm

Examples

library(voice)

url0 <- 'https://raw.githubusercontent.com/filipezabala/voiceAudios/main/rttm/sherlock0.rttm'
destfile0 <- paste0(tempdir(), '/sherlock0.rttm')
download.file(url0, destfile = destfile0)
url1 <- 'https://raw.githubusercontent.com/filipezabala/voiceAudios/main/rttm/sherlock1.rttm'
destfile1 <- paste0(tempdir(), '/sherlock1.rttm')
download.file(url0, destfile = destfile1)

rttm <- voice::read_rttm(dirname(destfile0))
(er <- voice::enrich_rttm(rttm))
class(er)
lapply(er, class)