Skip to contents

Get Audio File Durations via WAV Header Reading

Source: R/batch-queries.R
get_durations_batch.Rd

Reads only the 44-byte WAV header to calculate duration, avoiding full file loading via `LongSound$from_file()$get_total_duration()`.

This is a **Tier 4 "Ultra"** function for batch DSI (Dysphonia Severity Index) calculations where file duration is the MPT (Maximum Phonation Time) component.

Usage

get_durations_batch(file_paths)

Arguments

file_paths

Character vector of .wav file paths

Value

Numeric vector of durations (seconds). Returns `NA` for files that cannot be read or are not valid WAV files.

Performance

This function avoids the overhead of the standard LongSound approach by:

  • Reading only the first 44-100 bytes of the WAV header

  • Avoiding memory allocation for audio samples

  • Skipping all Praat object construction

API Tier

This is a **Tier 4 "Ultra"** function. Tier 4 functions keep entire workflows in the C++ layer to minimize R<->C++ boundary crossings, returning only final scalar or simple vector results.

See also

[LongSound()] for full audio file access when you need more than duration

Examples

wav1 <- tempfile(fileext = ".wav")
wav2 <- tempfile(fileext = ".wav")
Sound$create_tone(frequency = 150, duration = 0.3, sampling_rate =
 16000)$save(wav1)
Sound$create_tone(frequency = 150, duration = 0.5, sampling_rate =
 16000)$save(wav2)

# Multiple files (DSI workflow)
durations <- get_durations_batch(c(wav1, wav2))
max_mpt <- max(durations, na.rm = TRUE)