This browser is missing features the app needs (DecompressionStream, regex lookbehind, module workers). Please use a recent version of Chrome, Firefox, Edge, or Safari (16.4+).

text2IPA

Clean written text, generate broad IPA transcriptions, and extract phonological information with the Fonology R package.

English, French, Italian, Portuguese, and Spanish are supported.

Quick mode

Paste or type text, then receive a copyable IPA passage and a downloadable word-to-IPA table.

Use Quick mode
Batch mode

Upload a UTF-8 .txt file, select phonological variables, and download the results as CSV.

Analyze a file
A note about fallback transcriptions: an asterisk (*) marks a form generated by fallback rules rather than a lexical lookup. Learn more about Fonology.
IPA passage

Enter text and select Transcribe.

Token results
Analysis summary

Upload a text file and select Analyze file. Files over 5 MB or 250,000 cleaned tokens cannot be analyzed. Use Fonology package in R instead.

Preview (first 100 rows)
About text2IPA

text2IPA converts cleaned written text into broad IPA using the Fonology package.

  • Languages: English, French, Italian, Portuguese, and Spanish.
  • * marks a rule-derived fallback transcription.
  • NA marks an unavailable value or a syllable/constituent that does not exist.
  • Cleaning lowercases text and removes punctuation, numbers, and empty tokens.
  • Batch output preserves token order and repeated words.

Uploads must be UTF-8 .txt files no larger than 5 MB or 250,000 cleaned tokens.

This page runs entirely in your browser: no text you enter or upload leaves your device.

To gauge how much the tool is used, the site records anonymous counts: which panel was opened, the language, mode, and number of words of each analysis, and the country the request came from. No text, transcription, or file name is ever transmitted, no cookies are set, and no IP address is stored.

How to cite

If you use text2IPA, please cite the Fonology package that powers its analyses: