Automatic spelling correction

Aspose.OCR for Python via .NET can automatically replace commonly misspelled words in recognition results with the correct ones. This functionality supports the following languages:

Value	Language
SpellCheckLanguage.CES	Czech
SpellCheckLanguage.DAN	Danish
SpellCheckLanguage.DEU	German
SpellCheckLanguage.NLD	Dutch
SpellCheckLanguage.ENG	English
SpellCheckLanguage.EST	Estonian
SpellCheckLanguage.FIN	Finnish
SpellCheckLanguage.FRA	French
SpellCheckLanguage.ITA	Italian
SpellCheckLanguage.LAV	Latvian
SpellCheckLanguage.LIT	Lithuanian
SpellCheckLanguage.POL	Polish
SpellCheckLanguage.POR	Portuguese
SpellCheckLanguage.RON	Romanian
SpellCheckLanguage.SLK	Slovak
SpellCheckLanguage.SLV	Slovene
SpellCheckLanguage.SPA	Spanish
SpellCheckLanguage.SWE	Swedish

Spell checking is done on the fly, either on output (get_spell_check_corrected_text() method) or on saving the results:

Output corrected results

# Instantiate Aspose.OCR API
api = AsposeOcr()
# Add image to the recognition batch
input = OcrInput(InputType.SINGLE_IMAGE)
input.add("source.png")
# Recognize the image
result = api.recognize(input)
# Print corrected results
correctedResult = results[0].get_spell_check_corrected_text(SpellCheckLanguage.ENG)
print(correctedResult)

Save corrected results

# Instantiate Aspose.OCR API
api = AsposeOcr()
# Add image to the recognition batch
input = OcrInput(InputType.SINGLE_IMAGE)
input.add("source.png")
# Recognize the image
results = api.recognize(input)
# Save recognition result
results[0].save("result.txt", SaveFormat.TEXT, true, SpellCheckLanguage.DEU)

Automatic spelling correction requires additional processing resources and is disabled by default.

Spelling correction of any text

You can also correct misspelled words in any text string using correct_spelling() method of AsposeOcr class:

# Instantiate Aspose.OCR API
api = AsposeOcr()
# Correct spelling in string
text = "Die schönste Jungfrau sitzet dort oben wunderbar"
correctedText = api.correct_spelling(text, SpellCheckLanguage.DEU)
print(correctedText)