# text

## R Language Analysis Suite

An R-package for analyzing natural language with transformers-based
large language models. The `text` package is part of the *R Language
Analysis Suite*, including:

- [`talk`](https://www.r-talk.org/) - a package that transforms voice
  recordings into text, audio features, or embeddings.  
    
- [`text`](https://www.r-text.org/) - a package that provides tools for
  many language tasks such as converting digital text into word
  embeddings.  
    
  `talk` and `text` offer access to Large Language Models from Hugging
  Face.  
    
- [`topics`](https://www.r-topics.org/) a package with tools for
  visualizing language patterns into topics.  
    
- [`the L-BAM Library`](https://r-text.org/articles/LBAM.html) a library
  that provides pre-trained models for different psychological
  assessments such as mental health issues, personality and related
  behaviours.  
    

![](reference/figures/talk_text_topics.svg)

  
The *R Language Analysis Suite* is created through a collaboration
between psychology and computer science to address research needs and
ensure state-of-the-art techniques. The suite is continuously tested on
Ubuntu, Mac OS and Windows using the latest stable R version.

The *text*-package has two main objectives:  
\* First, to serve R-users as a *point solution* for transforming text
to state-of-the-art word embeddings that are ready to be used for
downstream tasks. The package provides a user-friendly link to language
models based on transformers from [Hugging
Face](https://huggingface.co/).  
\* Second, to serve as an *end-to-end solution* that provides
state-of-the-art AI techniques tailored for social and behavioral
scientists.  
Please reference our tutorial article when using the `text` package:
[The text-package: An R-package for Analyzing and Visualizing Human
Language Using Natural Language Processing and Deep
Learning](https://pubmed.ncbi.nlm.nih.gov/37126041/).  

> ### 📣 Online Workshop: *Analysing Human Language using R*
>
> **August 11–13, 2026** — three half-day sessions (2:00–5:00 pm CEST /
> 8:00–11:00 am ET)
>
> Learn the full *R Language Analysis Suite* —
> [`talk`](https://www.r-talk.org/), [`text`](https://www.r-text.org/),
> and [`topics`](https://www.r-topics.org/) — with Oscar Kjell.
>
> **[Read more & register →](https://smart-workshops.com/lang-r-info)**

### Point solution for transforming text to embeddings

Recent significant advances in NLP research have resulted in improved
representations of human language (i.e., language models). These
language models have produced big performance gains in tasks related to
understanding human language. Text are making these SOTA models easily
accessible through an interface to
[HuggingFace](https://huggingface.co/docs/transformers/index) in Python.

*Text* provides many of the contemporary state-of-the-art language
models that are based on deep learning to model word order and context.
Multilingual language models can also represent several languages;
multilingual BERT comprises *104 different languages*.

*Table 1. Some of the available language models*

| Models | References | Layers | Dimensions | Language |
|:---|:---|:---|:---|:---|
| ‘bert-base-uncased’ | [Devlin et al. 2019](https://aclanthology.org/N19-1423/) | 12 | 768 | English |
| ‘roberta-base’ | [Liu et al. 2019](https://arxiv.org/abs/1907.11692) | 12 | 768 | English |
| ‘distilbert-base-cased’ | [Sahn et al., 2019](https://arxiv.org/abs/1910.01108) | 6 | 768 | English |
| ‘bert-base-multilingual-cased’ | [Devlin et al. 2019](https://aclanthology.org/N19-1423/) | 12 | 768 | [104 top languages at Wikipedia](https://meta.wikimedia.org/wiki/List_of_Wikipedias) |
| ‘xlm-roberta-large’ | [Liu et al](https://arxiv.org/pdf/1907.11692) | 24 | 1024 | [100 language](https://huggingface.co/docs/transformers/multilingual) |

See [HuggingFace](https://huggingface.co/models/) for a more
comprehensive list of models.

### An end-to-end package

*Text* also provides functions to analyse the word embeddings with
well-tested machine learning algorithms and statistics. The focus is to
analyze and visualize text, and their relation to other text or
numerical variables. For example, the
[`textTrain()`](https://r-text.org/reference/textTrain.md) function is
used to examine how well the word embeddings from a text can predict a
numeric or categorical variable. Another example is functions plotting
statistically significant words in the word embedding space.

# Package index

## Installation

- [`textrpp_install()`](https://r-text.org/reference/textrpp_install.md)
  [`textrpp_install_virtualenv()`](https://r-text.org/reference/textrpp_install.md)
  : Install text required python packages in conda or virtualenv
  environment
- [`textrpp_uninstall()`](https://r-text.org/reference/textrpp_uninstall.md)
  : Uninstall textrpp conda environment
- [`textrpp_initialize()`](https://r-text.org/reference/textrpp_initialize.md)
  : Initialize text required python packages
- [`textDiagnostics()`](https://r-text.org/reference/textDiagnostics.md)
  : Run diagnostics for the text package

## Transform text to word embeddings

- [`textEmbed()`](https://r-text.org/reference/textEmbed.md) :
  textEmbed() extracts layers and aggregate them to word embeddings, for
  all character variables in a given dataframe.
- [`textDimName()`](https://r-text.org/reference/textDimName.md) :
  Change dimension names
- [`textEmbedRawLayers()`](https://r-text.org/reference/textEmbedRawLayers.md)
  : Extract layers of hidden states
- [`textEmbedLayerAggregation()`](https://r-text.org/reference/textEmbedLayerAggregation.md)
  : Aggregate layers
- [`textEmbedReduce()`](https://r-text.org/reference/textEmbedReduce.md)
  : Pre-trained dimension reduction (experimental)
- [`textEmbedStatic()`](https://r-text.org/reference/textEmbedStatic.md)
  : Apply static word embeddings

## Fine-tuning models

- [`textFineTuneTask()`](https://r-text.org/reference/textFineTuneTask.md)
  : Task Adapted Pre-Training (EXPERIMENTAL - under development)
- [`textFineTuneDomain()`](https://r-text.org/reference/textFineTuneDomain.md)
  : Domain Adapted Pre-Training (EXPERIMENTAL - under development)

## Text language analysis tasks

- [`textGeneration()`](https://r-text.org/reference/textGeneration.md) :
  Text generation
- [`textNER()`](https://r-text.org/reference/textNER.md) : Named Entity
  Recognition. (experimental)
- [`textSum()`](https://r-text.org/reference/textSum.md) : Summarize
  texts. (experimental)
- [`textQA()`](https://r-text.org/reference/textQA.md) : Question
  Answering. (experimental)
- [`textTranslate()`](https://r-text.org/reference/textTranslate.md) :
  Translation. (experimental)
- [`textZeroShot()`](https://r-text.org/reference/textZeroShot.md) :
  Zero Shot Classification (Experimental)

## The text-train functions

- [`textTrain()`](https://r-text.org/reference/textTrain.md) : Trains
  word embeddings
- [`textTrainLists()`](https://r-text.org/reference/textTrainLists.md) :
  Train lists of word embeddings
- [`textTrainRegression()`](https://r-text.org/reference/textTrainRegression.md)
  : Train word embeddings to a numeric variable.
- [`textTrainRandomForest()`](https://r-text.org/reference/textTrainRandomForest.md)
  : Trains word embeddings usig random forest
- [`textTrainN()`](https://r-text.org/reference/textTrainN.md) :
  Cross-validated accuracies across sample-sizes
- [`textTrainNPlot()`](https://r-text.org/reference/textTrainNPlot.md) :
  Plot cross-validated accuracies across sample sizes
- [`textExamples()`](https://r-text.org/reference/textExamples.md) :
  Identify language examples.

## The text-predict functions

- [`textPredict()`](https://r-text.org/reference/textPredict.md)
  [`textAssess()`](https://r-text.org/reference/textPredict.md)
  [`textClassify()`](https://r-text.org/reference/textPredict.md) :
  textPredict, textAssess and textClassify
- [`textPredictTest()`](https://r-text.org/reference/textPredictTest.md)
  : Significance testing for model prediction performance
- [`textPredictAll()`](https://r-text.org/reference/textPredictAll.md) :
  Predict from several models, selecting the correct input
- [`textLBAM()`](https://r-text.org/reference/textLBAM.md) : The LBAM
  library

## Semantic similarities and distances functions

- [`textSimilarity()`](https://r-text.org/reference/textSimilarity.md) :
  Semantic Similarity
- [`textDistance()`](https://r-text.org/reference/textDistance.md) :
  Semantic distance
- [`textSimilarityMatrix()`](https://r-text.org/reference/textSimilarityMatrix.md)
  : Semantic similarity across multiple word embeddings
- [`textDistanceMatrix()`](https://r-text.org/reference/textDistanceMatrix.md)
  : Semantic distance across multiple word embeddings
- [`textSimilarityNorm()`](https://r-text.org/reference/textSimilarityNorm.md)
  : Semantic similarity between a text variable and a word norm
- [`textDistanceNorm()`](https://r-text.org/reference/textDistanceNorm.md)
  : Semantic distance between a text variable and a word norm

## Visualise words in the word embedding space

- [`textProjection()`](https://r-text.org/reference/textProjection.md) :
  Supervised Dimension Projection
- [`textWordPrediction()`](https://r-text.org/reference/textWordPrediction.md)
  : Compute word-level prediction scores for plotting with
  textProjectionPlot().
- [`textPlot()`](https://r-text.org/reference/textPlot.md) : Plot words
- [`textProjectionPlot()`](https://r-text.org/reference/textProjectionPlot.md)
  : Plot Supervised Dimension Projection
- [`textCentrality()`](https://r-text.org/reference/textCentrality.md) :
  Semantic similarity score between single words' and an aggregated word
  embeddings
- [`textCentralityPlot()`](https://r-text.org/reference/textCentralityPlot.md)
  : Plots words from textCentrality()
- [`textPCA()`](https://r-text.org/reference/textPCA.md) : textPCA()
- [`textPCAPlot()`](https://r-text.org/reference/textPCAPlot.md) :
  textPCAPlot

## BERTopics

- [`textTopics()`](https://r-text.org/reference/textTopics.md)
  [`textBERTopics()`](https://r-text.org/reference/textTopics.md) :
  BERTopic
- [`textTopicsTest()`](https://r-text.org/reference/textTopicsTest.md) :
  Wrapper for topicsTest function from the topics package
- [`textTopicsWordcloud()`](https://r-text.org/reference/textTopicsWordcloud.md)
  : Plot word clouds
- [`textTopicsReduce()`](https://r-text.org/reference/textTopicsReduce.md)
  : textTopicsReduce (EXPERIMENTAL)
- [`textTopicsTree()`](https://r-text.org/reference/textTopicsTree.md) :
  textTopicsTest (EXPERIMENTAL) to get the hierarchical topic tree

## View or delete downloaded HuggingFace models

- [`textModels()`](https://r-text.org/reference/textModels.md) : Check
  downloaded, available models.
- [`textModelLayers()`](https://r-text.org/reference/textModelLayers.md)
  : Number of layers
- [`textModelsRemove()`](https://r-text.org/reference/textModelsRemove.md)
  : Delete a specified model

## Miscellaneous

- [`textDescriptives()`](https://r-text.org/reference/textDescriptives.md)
  : Compute descriptive statistics of character variables.
- [`textClean()`](https://r-text.org/reference/textClean.md) : Cleans
  text from standard personal information
- [`textTokenize()`](https://r-text.org/reference/textTokenize.md) :
  Tokenize text-variables
- [`textTokenizeAndCount()`](https://r-text.org/reference/textTokenizeAndCount.md)
  : Tokenize and count
- [`textDomainCompare()`](https://r-text.org/reference/textDomainCompare.md)
  : Compare two language domains
- [`textFindNonASCII()`](https://r-text.org/reference/textFindNonASCII.md)
  : Detect non-ASCII characters
- [`textCleanNonASCII()`](https://r-text.org/reference/textCleanNonASCII.md)
  : Clean non-ASCII characters

## Example Data

- [`Language_based_assessment_data_8`](https://r-text.org/reference/Language_based_assessment_data_8.md)
  : Text and numeric data for 10 participants.
- [`word_embeddings_4`](https://r-text.org/reference/word_embeddings_4.md)
  : Word embeddings for 4 text variables for 40 participants
- [`raw_embeddings_1`](https://r-text.org/reference/raw_embeddings_1.md)
  : Word embeddings from textEmbedRawLayers function
- [`Language_based_assessment_data_3_100`](https://r-text.org/reference/Language_based_assessment_data_3_100.md)
  : Example text and numeric data.
- [`DP_projections_HILS_SWLS_100`](https://r-text.org/reference/DP_projections_HILS_SWLS_100.md)
  : Data for plotting a Dot Product Projection Plot.
- [`centrality_data_harmony`](https://r-text.org/reference/centrality_data_harmony.md)
  : Example data for plotting a Semantic Centrality Plot.
- [`PC_projections_satisfactionwords_40`](https://r-text.org/reference/PC_projections_satisfactionwords_40.md)
  : Example data for plotting a Principle Component Projection Plot.

# Articles

### Resources

- [Resources](https://r-text.org/articles/resources.md):

### Set-up

- [The Language-Based Assessment Model (L-BAM)
  Library](https://r-text.org/articles/LBAM.md):
- [Extended Installation
  Guide](https://r-text.org/articles/ext_install_guide.md):
- [Installing and Managing Python Environments with
  \`reticulate\`](https://r-text.org/articles/reticulate.md):
- [Creating a Singularity Container to Run HuggingFace Transformers
  Models in
  R](https://r-text.org/articles/singularity_transformers_container.md):
- [How to best manage computationally heavy
  analyses](https://r-text.org/articles/huggingface_in_r_and_computer_capacity.md):
- [HuggingFace language models are downloaded in
  .cache](https://r-text.org/articles/removing_huggingface_transformers_cache_files.md):

### Tutorials

These tutorials go into detail about various text-analyses.

- [HuggingFace Transformers in R: Word Embeddings Defaults and
  Specifications](https://r-text.org/articles/huggingface_in_r.md):
- [Psychological Methods: the Text
  Tutorial](https://r-text.org/articles/psychological_methods.md):
- [L-BAM Tutorial](https://r-text.org/articles/lbam_tutorial.md):
- [Implicit Motives
  Tutorial](https://r-text.org/articles/implicit_motives_tutorial.md):

### Comments

General comments about text-analyses.

- [Pre-registration and Researcher Degrees of
  Freedom](https://r-text.org/articles/pre_registration_and_transformers.md):