This page explains what Phrase Search does with TV dialogue, how much of it is ever shown, and how to ask for a title to be removed.
Phrase Search is a language-analysis tool for English learners. You enter a phrase and it reports how that phrase behaves across a corpus of TV dialogue — how often it occurs, which word forms it takes, which words it appears with, and how its use varies by genre.
The output is statistics. Individual lines of dialogue appear only as evidence supporting those statistics, and only in short extracts.
Dialogue is drawn from the OpenSubtitles corpus published by the OPUS project (P. Lison and J. Tiedemann, 2016, LREC), a dataset distributed for language research. Series and episode metadata comes from IMDb. Full source credits appear in the footer of every page.
The titles included were selected individually rather than harvested indiscriminately, and the corpus covers a limited set of English-language series.
If you hold rights in a title and want it removed, please get in touch. You do not need to use any particular form of words. It helps if you can tell us:
Removal is handled per title. When a title is removed, its dialogue is deleted from the index, and it stops appearing in search results, statistics, and every phrase page.
We aim to acknowledge requests within 7 days. Requests are acted on rather than argued with: if you would prefer a title not to appear here, we will take it out.
Use the Contact link at the bottom of any page, including this one. It opens a short form and reaches us directly.