Extract Pitch (AC) from Multiple Sounds in Single C++ Call — sound_to_pitch_ac

Batch version of to_pitch_ac with full voicing parameters. Avoids O(n) R→C boundary crossings for VUV analysis workflows.

Usage

sound_to_pitch_ac_batch(
  sounds,
  time_step = 0,
  pitch_floor = 75,
  pitch_ceiling = 600,
  max_candidates = 15L,
  very_accurate = FALSE,
  silence_threshold = 0.03,
  voicing_threshold = 0.45,
  octave_cost = 0.01,
  octave_jump_cost = 0.35,
  voiced_unvoiced_cost = 0.14,
  return_r6 = TRUE
)

Arguments

sounds: List of Sound objects (R6) or external pointers
time_step: Numeric. Time step (0 = automatic)
pitch_floor: Numeric. Pitch floor in Hz (default: 75)
pitch_ceiling: Numeric. Pitch ceiling in Hz (default: 600)
max_candidates: Integer. Max candidates per frame (default: 15)
very_accurate: Logical. Use very accurate algorithm (default: FALSE)
silence_threshold: Numeric. Silence threshold (default: 0.03)
voicing_threshold: Numeric. Voicing threshold (default: 0.45)
octave_cost: Numeric. Octave cost (default: 0.01)
octave_jump_cost: Numeric. Octave jump cost (default: 0.35)
voiced_unvoiced_cost: Numeric. Voiced/unvoiced cost (default: 0.14)
return_r6: Logical. Return R6 Pitch objects (TRUE) or raw xptrs (FALSE)

Value

List of Pitch objects (R6 or xptr depending on return_r6)

Examples

sounds <- list(
  Sound$create_tone(frequency = 150, duration = 0.5),
  Sound$create_tone(frequency = 200, duration = 0.5)
)
pitches <- sound_to_pitch_ac_batch(sounds)