Effective date: September 12, 2026
This document is provided in English, which is the authoritative version.
DolphinUp is built on open data and open-source software created by other people. This page credits those sources, states the licence each one is used under, and explains what we changed. Where a licence requires it, the licence text is linked and the corresponding part of our data is offered back under the same licence. If you believe something here is incomplete or incorrect, please write to support@dolphinup.com and we will fix it.
The collection "Từ vựng luyện thi TOEIC®" teaches a set of words chosen from the TOEIC Service List, a published research word list compiled by Charles Browne and Brent Culligan and released under the Creative Commons Attribution-ShareAlike licence. We use it to decide which words the collection teaches and in what order.
Every Vietnamese meaning and every example sentence in that collection is written by DolphinUp. The source supplies the choice of words, nothing that you read.
In line with the ShareAlike condition, the word list we derived — the selection and its ordering — is itself offered under the Creative Commons Attribution-ShareAlike 4.0 International licence. Our example sentences and meanings are separate, independently written works and are not covered by that licence.
TOEIC is a registered trademark of ETS. DolphinUp is not affiliated with, endorsed by, or approved by ETS. We use the name only to describe what the collection prepares you for. The collection contains no ETS logo and no real test questions or official ETS material; the word list is independent research about the language the exam uses.
Word meanings, parts of speech and phonetic transcriptions for most language pairs come from Wiktionary, via the machine-readable Wiktextract dataset published at kaikki.org. Wiktionary is written by volunteers and published under the Creative Commons Attribution-ShareAlike licence.
We do not present Wiktionary as it stands. Our dictionary is an adapted version: entries are filtered to English headwords, definitions are shortened and limited in number, senses are regrouped by part of speech, headwords are normalised, and meanings from other sources are merged in where Wiktionary has none. Errors you find in DolphinUp's dictionary are therefore ours, not Wiktionary's.
In line with the ShareAlike condition, the adapted dictionary data that DolphinUp makes available through the app is itself offered under the Creative Commons Attribution-ShareAlike 4.0 International licence. This applies to the dictionary data only; the DolphinUp application, its interface, its lessons and its other content are separate works and are not covered by that licence.
For English→Arabic and English→Hindi, the meanings come from the FreeDict project instead: the Arabic data originates in the Arabeyes Wordlist project and the Hindi data in the Anusaaraka English-Hindi dictionary. Both are published under a copyleft licence of the GNU General Public Licence family. As with Wiktionary, the data has been reformatted, filtered and shortened for use in DolphinUp.
Where the Wiktionary edition for your language never described a word, DolphinUp falls back to a translation from PanLex, a large aggregation of bilingual dictionaries released into the public domain under CC0. PanLex is used only to fill gaps: it never replaces a real definition and never adds words of its own.
Where Wiktionary carries no phonetic transcription, DolphinUp uses the CMU Pronouncing Dictionary (CMUdict) from Carnegie Mellon University, converted from its ARPAbet notation into the same IPA notation used everywhere else in the app. CMUdict is distributed under the two-clause BSD licence, which requires the following notice to be reproduced:
Copyright (C) 1993-2015 Carnegie Mellon University. All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: (1) Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer; the contents of that file are deemed to be source code. (2) Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
This work was supported in part by funding from the Defense Advanced Research Projects Agency, the Office of Naval Research and the National Science Foundation of the United States of America, and by member companies of the Carnegie Mellon Sphinx Speech Consortium. We acknowledge the contributions of many volunteers to the expansion and improvement of this dictionary.
THIS SOFTWARE IS PROVIDED BY CARNEGIE MELLON UNIVERSITY "AS IS" AND ANY EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL CARNEGIE MELLON UNIVERSITY NOR ITS EMPLOYEES BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
For rare and technical words covered by neither source above, an approximate pronunciation is generated with the open-source speech synthesiser espeak-ng (GNU GPL v3). Generation happens when our dictionary is built; espeak-ng is not part of the running app.
Recorded pronunciations are played directly from Wikimedia Commons; we do not copy or re-host them. Each recording is a separate work with its own author and licence — most are Creative Commons Attribution-ShareAlike or CC0, and many come from the Lingua Libre recording project. Because the credit differs from file to file, every pronunciation in DolphinUp links to that recording's own page on Wikimedia Commons, where its author and licence are stated. To open it: on a computer, use the ⓘ next to the play button; on a phone, press and hold the play button. A plain tap plays the sound instead.
DolphinUp ships no background images of its own. Instead you can search for one from inside the app and apply it in a click. That search runs through Openverse, which records the licence of every photo, and DolphinUp restricts it to photos anyone may use freely: public-domain works and photos whose only condition is that their author be credited. Photos flagged as mature are excluded.
The photo you pick is used from its original location — DolphinUp does not copy it — and its credit line is shown in your appearance settings for as long as it is your background. If you upload your own image instead, it stays on your device and is not sent to us.
Depending on the collection, the words taught are chosen either by how often they actually occur in everyday speech, or from a published research word list. In both cases we use openly-licensed lists that permit this use and do not use proprietary or unlicensed word lists.
Some collections order vocabulary by how often words actually occur in everyday speech, derived from an open frequency list built from the OpenSubtitles corpora and released under the Creative Commons Attribution-ShareAlike licence. Only the statistical ordering is used — no text from that corpus is shown in the app. In line with the ShareAlike condition, the word list we derived from it — the selection and its ordering — is itself offered under the Creative Commons Attribution-ShareAlike 4.0 International licence. Our example sentences and meanings are separate, independently written works and are not covered by that licence.
Another collection teaches a set of words combined from five published research word lists — the New General Service List, NGSL-GR, the Business Service List, the TOEIC Service List and the New Academic Word List — compiled by Charles Browne, Brent Culligan and Joseph Phillips and released under the Creative Commons Attribution-ShareAlike licence. (The TOEIC Service List is also used on its own in the TOEIC collection — see "TOEIC Vocabulary Selection" above.) We filtered, combined and reordered these five lists for teaching; none of their source text is shown to you — every meaning and example sentence is written by DolphinUp. In line with the ShareAlike condition, the combined word list we derived is itself offered under the Creative Commons Attribution-ShareAlike 4.0 International licence. Our example sentences and meanings are separate, independently written works and are not covered by that licence.
The same project's lists also power the "Vocabulary set" filter in the dictionary, which lets you narrow the word list you are browsing. That filter uses the five lists above plus three more from the same authors: the NGSL frequency table of 31,240 measured words, NGSL-Spoken and the New Dolch List. We record which of our dictionary entries each list names, so choosing a set shows exactly the words that list publishes. That membership data is likewise offered under the Creative Commons Attribution-ShareAlike 4.0 International licence.
Video lessons are played through YouTube's own embedded player. The videos belong to the channels that published them, and a video can only be added to DolphinUp by someone able to label it on YouTube for this purpose. DolphinUp does not sell access to YouTube content: video lessons can be watched without an account and are never counted against the free daily learning time. Your use of the embedded player is also subject to YouTube's own terms. YouTube and the YouTube logo are trademarks of Google LLC; DolphinUp is not affiliated with, sponsored by, or endorsed by Google.
Spoken pronunciation of sentences and words is produced by the speech voices available on your device: on the web, the voices your browser offers; in the DolphinUp apps for iPhone, iPad and Android, the voices of the device's operating system. Depending on the voice, the audio is generated either on your device or by the online speech service of the company that provides that voice (for example Google or Microsoft). When an online voice is used, the text being spoken is sent to that company so it can produce the audio. DolphinUp does not record, store or redistribute the audio.
DolphinUp is built with open-source software, used under permissive licences (MIT, Apache 2.0, BSD and similar). Among the larger components:
Google, YouTube, Microsoft, Edge, Chrome, Apple, Safari, Paddle, Cloudflare and all other names and logos referred to in the app are the trademarks of their respective owners. They are used only to identify those products and services. DolphinUp is an independent product and is not affiliated with, sponsored by, or endorsed by any of them.
If you are the rights holder of anything used here and the credit is wrong, missing, or you would like the material removed, contact us at support@dolphinup.com. We take these requests seriously and act on them promptly. Copyright complaints about content posted by users of the Service are handled under the "Copyright Complaints" section of our Terms of Service.