Vocabsieve

Latest version: v0.12.4

Safety actively analyzes 701298 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 2 of 6

0.10.2

New features
- Option to capitalize sentences from clipboard (by AtilioA)
- Async audio fetching for better UI responsiveness (by AtilioA)

Bug fixes
- Fetching all pronunciations (by artjomsR)
- KOReader importer should now truncate language codes with regional suffixes. (by billyc)

Other changes
- Default instance of Lingva translate is now https://lingva.lunar.icu. Existing users will need to change this manually.

0.10.1

Users upgrading from before 0.10 should also view the release notes of 0.10.0. It is NOT safe to downgrade to a version before 0.10 after upgrading due to a database change.
New features
- Implemented basic dark theme support. The theme should also look better overall especially for Windows users.

Bug fixes
- Kindle importer should now record the correct lookup timestamps. For existing users a database change will take place to fix this.
- KOReader vocab builder importer should now display a correct count
- Python package should now have all dependencies listed

0.10.0

It is NOT safe to downgrade to a version before 0.10 after upgrading due to a database change.
New features
- Vocabulary tracking. From this version, VocabSieve is now able to process data from a variety of sources, including Anki, past lookups, and texts you have read (incl. video subtitles) to create a database of words you know. Parameters are configurable via the settings.
- You can export either the exact data or simply a list of known words in JSON format for your own use.
- Book analyzer. Based on the words you know, a book analyzer is provided to help you quickly screen for immersion content based on difficulty and ease of making flashcards.
- The learning simulator takes into account effects of learning words on text difficulty over the course of reading.
- Statistics. Several statistics are now viewable in the program itself, via the Statistics option on the menu bar.
- A new KOReader importer based on its Vocabulary Builder is available. Due to the various limitations and inherent complexity of the original KOReader importer, it is now considered deprecated, and is planned to be removed two releases from now (in 0.12). To use it, please ensure that your KOReader vocabulary builder is set to save context, and not add words automatically to reduce noise.
- Cognates are now supported as a type of dictionary. It can help make vocabulary tracking more accurate, especially if your target language shares a lot of words in common with languages you know.
Other changes
- When multiple words are selected, it will now brute-force through all combinations of lemmatized and non-lemmatized versions of each word until a definition is found. This should improve dictionary usability with phrases.
- You can now use compressed json formats (.json.xz, .json.gz, .json.bz2). NOT zip, tar, 7z, or .json.zst. It should behave the same as if you first decompress the json.
- Kindle and KOReader imports will now import the entirety of your lookup histories automatically, before selecting any book to make flashcards from. This provides a fuller picture of the words you looked up.
- KOReader users (regardless of using the old or the new importer) should now select the directory containing both the KOReader settings folder and the books. For most users, the easiest way to do this is to select the device directly for Kobo or the user's home folder (`/storage/emulated/0`) for Android.

0.9.2

Fixes
- Define and Define (direct) buttons should work as it did in v0.8.3.

Changes
- Importers will now remember the latest note from your eReader. This should prevent annoyances where you did not synchronize your notes properly, or if your eReader clock is wrong.
- The Windows/Mac packages should now support PyMorphy2. You can expect a large lemmatization performance improvement for Russian and Ukrainian.

Note: Windows users will see a console window when opening the program. This is due to a mistake in the build process. It does not indicate a problem. Please simply ignore it for now. It will be fixed in the next release.

0.9.1

This version fixes an error affecting some users upgrading from earlier versions to v0.9.0.

0.8.3

This version fixes a bug that causes the program to not launch in the previous release (0.8.2) . Apologies for the delay.

Page 2 of 6

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.