Function based on original from the signglossR package. Reads either all ELAN files (.eaf) from a directory (if path is set to a directory) or an individual file (if path is set to a file). Returns a tibble dataframe with all the data.
Example:
# Read all the EAF files of the directory into a single tibble dataframe
all_eafs <- read_eaf(path="/Users/username/Desktop/Corpus/EAFs/")
# Read an individual EAF file into a single tibble dataframe
my_eaf <- read_eaf("/Users/username/Desktop/Corpus/EAFs/FrogStory_2022-12-01.eaf")
Uses the read_eaf() function to first read an ELAN file (or all ELAN files in a directory) and then split one (if single file path) or more (if entire directory) video files into clips corresponding to the segments on a tier in the ELAN file running ffmpeg commands. The tier used for segmentation needs to be provided. Default input and output video file format is .mp4 but can be specified with the arguments. Padding can be added with positive or negative values (in milliseconds) adding/removing frames at the beginning and end of the individual clips relative to the cell segment durations.
Usage:
split_elan_video(elan_path = "/path/to/eaf/file(s)",
segmentation_tier = "name_of_segmentation_tier",
video_path = "/path/to/video/file(s)",
annotation_tag = T, # will add contents of ELAN cells in output filenames
padding = 0, # adds (or subtracts if negative) frames (in milliseconds) before+after segment duration
video_input_format = ".mov", # specify input video format in directory (default is .mp4)
video_output_format = ".mp4") # specify output video format (default is .mp4)
Example:
# Split all videos (.mov) with file names mirroring the EAF file names and output .mp4 files
split_elan_video(elan_path = "/Users/username/Desktop/Corpus/EAFs/",
segmentation_tier = "name_of_segmentation_tier",
video_path = "/Users/username/Desktop/Corpus/videos/",
annotation_tag = T,
video_input_format = ".mov",
video_output_format = ".mp4")
# Split a specific video based on an EAF file with a different name, adding 100 milliseconds before and after each segment
split_elan_video(elan_path = "/Users/username/Desktop/Corpus/EAFs/FrogStory_2022-12-01.eaf",
segmentation_tier = "name_of_segmentation_tier",
video_path = "/Users/username/Desktop/Corpus/videos/FrogStory_Signer012.mp4",
annotation_tag = T,
padding = 100)
Lets you merge sequences of annotations (in an ELAN type format) into longer utterances ("turns") based on the chronological order of the annotations' start times within each file.
Lets you merge sequences of annotations (in an ELAN type format) into longer utterances ("turns") based on the chronological order of the annotations' start times and a defined interval of maximum allowed duration to the next annotation, by participant (e.g., signer/speaker) within each file.
Lets you merge sequences of annotations (in an ELAN type format) into longer utterances ("turns") based on the participant who is uttering the most within each window of 3 annotations chronologically based on the annotations' start times within each file.
The file pacific_map.R contains a simple outline for a Pacific-centered basemap using ggplot2. Pacific-centered maps are often preferred in e.g. linguistics.
See also Hedvig Skirgårds solution for Pacific-centered (language) maps in R: HedvigS: Worldmap plotting