With release v5.7.0 we introduced an enhanced AI-based proofreading engine for English. It was built using modern approaches in Natural Language Processing (NLP). Unlike the classical rules or dictionary-based engines, it's based on the huge pre-trained language model. Its accuracy is at least 3 times better.
However, at the same time, there are some cons that should be taken into consideration:
- It is available for general English. It doesn't distinguish dialects. Thus, offered as a separate language.
- Suggestions can be unpredictable and there is no option to ignore a rule.
- AI technology consumes more resources to provide the results. It's response time is longer compared to classical engines. Check the updated WebSpellChecker Server v5.7.0 hardware requirements.
- It is available only for modern products, such as the WProofreader add-on for rich text editors and API.
- There are no tips or differentiation by problem type (spelling, grammar, style, punctuation). All are combined in one. Later we will add types.
You can give it a try using one of the public demos on our website: https://webspellchecker.com/wsc-proofreader/
If you are using the server-based package of WebSpellChecker v5.7.0 and higher, then you have an AI engine included in the package.
<Language Id="en_AI"> <Alias>ai</Alias> <GrammarCheckProvider>neural</GrammarCheckProvider> <GrammarCheckProviderOptions>en</GrammarCheckProviderOptions> <!--<ThesaurusEnabled>true</ThesaurusEnabled> <SpellEngineOptions> <Locale>am</Locale> <SpellCheckProvider>ssce</SpellCheckProvider> <Dictionary FullPath="ssceam2.clx" > <ForSuggest>no</ForSuggest> </Dictionary> <Dictionary FullPath="ssceam2s.clx"> <ForSuggest>yes</ForSuggest> </Dictionary> <Dictionary FullPath="keywords.clx"></Dictionary> <Dictionary FullPath="ssceam.tlx"></Dictionary> </SpellEngineOptions>--> </Language>