multimolang is a package designed to support multimodal data analysis for linguistic research. It serves as the main repository for the MULTIFLOW project and provides tools to process, analyze, and derive insights from multimodal datasets, starting with gesture trajectories extracted from video data.

The first implemented tool, dfMaker, processes raw OpenPose data into structured dataframes, optimized for large-scale, multimodal data science workflows. Future versions of multimolang will include additional tools for prosody analysis and multimodal language integration.

If you use multimolang or any of its core functions (such as dfMaker) in your research, please cite:

Herreño Jiménez, B. (2024). multimolang: Multimodal Language Analysis [R package version 0.1.1]. The R Foundation. https://doi.org/10.32614/cran.package.multimolang

You can also retrieve the BibTeX entry by running in R:

citation("multimolang")

# Install from CRAN
install.packages("multimolang")

# Or install using devtools
devtools::install_github("daedalusLAB/multimolang")

For detailed installation instructions and usage examples, please refer to the package vignettes.

Currently, the main function in the package is dfMaker, which provides:

• Flexible Input Configuration: Accepts JSON files from OpenPose as input.
• Transformation Capabilities: Scales and normalizes body keypoints for gesture trajectory analysis.
• Big Data Optimization: Designed to handle large-scale multimodal datasets efficiently.

# Define paths to example data included with the package
input.folder <- system.file("extdata", "ex_videos", "o1", package = "multimolang")
output.file <- file.path(tempdir(), "processed_data.csv")
output.path <- tempdir()

# Run dfMaker with example data
df <- dfMaker(
  input.folder = input.folder,
  output.file = output.file,
  output.path = output.path,
  no_save = FALSE,
  fast_scaling = TRUE,
  transformation_coords = c(1, 1, 5, 5)
)

# View the resulting data
head(df)

Below is the current progress and pending tasks for multimolang:

Note: Future updates will incorporate tools for prosody and voice analysis, as well as co-speech gesture processing.

The MULTIFLOW project investigates the interplay between gesture, prosody, and language to uncover multimodal signatures in communication. Learn more about our research and goals.

• R functions for creating dataframes from OpenPose raw data.
  • Analyzing gesture trajectories from normalized body keypoint detection.
  • Building Massive Co-Speech Gesture Datasets for Specific Linguistic Patterns.
  • Gestural behavior is systematically attuned to language: Novel data analysis of co-speech gesture and its implications for multimodal interfaces.

We welcome contributions to multimolang! For bug reports, feature requests, or collaboration inquiries, open an issue on GitHub.

• dfMaker Details

• Updated file naming conventions to comply with CRAN requirements.
• Improved compatibility with CRAN by refining documentation and dependencies.
• Minor fixes to dfMaker for enhanced error handling and performance.
• Added package vignettes for detailed usage examples.

For a complete list of changes, refer to the CHANGELOG in the repository.