Batch merging using Praat's O(n) algorithm instead of O(n²) manual tier copying. Manual merge requires save/reload + insert_boundary for each interval (each insert shifts all later intervals). Batch merge is single-pass.
Details
Manual merge requires save/reload plus an O(n²) sequence of `insert_boundary` calls (each insert shifts all later intervals); batch merge does a single O(n) pass with proper interval handling.
**Domain handling:** - If `equalize_domains = FALSE` (default): * New domain runs from min(xmin) to max(xmax) of all input TextGrids * Tiers retain their original domains
- If `equalize_domains = TRUE`: * All tiers extended to the new domain * Empty intervals added at edges if needed
**Use cases:** - VUV analysis: Merging original TextGrid with VUV tier - Multi-annotator: Combining annotations from different annotators - Workflow: Adding automatic tiers to manual annotations
See also
TextGrid for TextGrid object creation
Other batch-ops:
sound_load_window()
Examples
# Create test TextGrids
tg1 <- textgrid_create(0, 1, "words")
tg1$insert_boundary(1, 0.5)
tg1$set_interval_text(1, 1, "hello")
tg1$set_interval_text(1, 2, "world")
tg2 <- textgrid_create(0, 1, "events", "events")
tg2$insert_point(1, 0.25, "click")
# Batch merge
merged <- textgrid_merge(list(tg1, tg2))
# Result has 2 tiers: "words" (interval) + "events" (point)
# With domain equalization
merged_eq <- textgrid_merge(list(tg1, tg2), equalize_domains = TRUE)