CODEWITHSUNDEEP
azureazure-cognitive-servicesmicrosoft-translator
Błażej
Błażej

Reputation: 1

Can Microsoft Translator API translate text with special characters?

I am trying to use Microsoft Translator API to translate text from Polish to any other language. In Polish, there are a couple of special characters like "ą", "ś", "ż" etc. When I send the HTTP request with no special characters:

POST /translate?api-version=3.0&from=pl&to=en HTTP/1.1
Ocp-Apim-Subscription-Key: ********
Ocp-Apim-Subscription-Region: ******
Content-Length: 21
Host: api.cognitive.microsofttranslator.com
Connection: close
User-Agent: Apache-HttpClient/4.5.10 (Java/15.0.2)
Accept-Encoding: gzip, deflate

[{"Text": "Gramatyka"}]

I receive a correct translation:

[{"translations":[{"text":"grammar","to":"en"}]}]

However, it is likely that a Polish word or sentence contains special characters:

POST /translate?api-version=3.0&from=pl&to=en HTTP/1.1
Ocp-Apim-Subscription-Key: ********
Ocp-Apim-Subscription-Region: ********
Content-Length: 21
Host: api.cognitive.microsofttranslator.com
Connection: close
User-Agent: Apache-HttpClient/4.5.10 (Java/15.0.2)
Accept-Encoding: gzip, deflate

[{"Text": "Roślina"}]

This request results in error code 400000:

{"error":{"code":400000,"message":"One of the request inputs is not valid."}}

If I change the special characters to standard ones (like change "ś" into "s"), the API does not give a proper translation. For example:

[{"Text": "Roslina"}]

results in:

[{"translations":[{"text":"Roslina","to":"en"}]}]

Whereas "roślina" should translate to "plant".

This problem applies to other languages too. For example German:

[{"Text": "Wörterbuch"}]

results in an 400000 error as well.

Has anyone found a solution to this?

Upvotes: 0

Views: 824

Answers (1)

Biplab Sah
Biplab Sah

Reputation: 79

Did you try checking the language detect score, to just understand if it is taking it as Polish. Can you try without "From" attribute. Make sure you put all headers. curl -X POST "https://api.cognitive.microsofttranslator.com/translate?api-version=3.0&to=zh-Hans" -H "Ocp-Apim-Subscription-Key: " -H "Content-Type: application/json; charset=UTF-8" -d "[{'Text':'Hello, what is your name?'}]"

Upvotes: 0

Related Questions