Spell Checker in English Text

Description

Spell Checker is a sequence-to-sequence pipeline that detects and corrects spelling errors in your input text. It’s based on Levenshtein Automaton for generating candidate corrections and a Neural Language Model for ranking corrections. You can download the pretrained pipeline that comes ready to use.

Download Copy S3 URI

How to use

pipeline_local = PretrainedPipeline("check_spelling_dl")

testDoc = '''
During the summer we have the hottest ueather. I have a black ueather jacket, so nice.I intrduce you to my sister, she is called ueather.
'''

result=pipeline_local.annotate(testDoc)
import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline

val pipeline = new PretrainedPipeline("check_spelling_dl", lang = "en")
val result = pipeline.annotate("During the summer we have the hottest ueather. I have a black ueather jacket, so nice.I intrduce you to my sister, she is called ueather.")
import nlu
nlu.load("en.spell").predict("""
During the summer we have the hottest ueather. I have a black ueather jacket, so nice.I intrduce you to my sister, she is called ueather.
""")

Results

('During', 'During'),
('the', 'the'),
('summer', 'summer'),
('we', 'we'),
('have', 'have'),
('the', 'the'),
('hottest', 'hottest'),
('ueather', 'weather'),
('.', '.'),
('I', 'I'),
('have', 'have'),
('a', 'a'),
('black', 'black'),
('ueather', 'leather'),
('jacket', 'jacket'),
(',', ','),
('so', 'so'),
('nice', 'nice'),
('.', '.'),
('I', 'I'),
('intrduce', 'introduce'),
('you', 'you'),
('to', 'to'),
('my', 'my'),
('sister', 'sister'),
(',', ','),
('she', 'she'),
('is', 'is'),
('called', 'called'),
('ueather', 'Heather'),
('.', '.')

Model Information

Model Name: check_spelling_dl
Type: pipeline
Compatibility: Spark NLP 3.3.4+
License: Open Source
Edition: Official
Language: en
Size: 118.1 MB

Included Models

  • DocumentAssembler
  • SentenceDetectorDLModel
  • TokenizerModel
  • ContextSpellCheckerModel