Language Detection¶

Language detection can be called separately or as part of other functions.

Recognized languages ¶

Our default mode distinguishes 31 languages:

ar - Arabic	el - Greek	he - Hebrew	it - Italian	nl - Dutch	sk - Slovak	zh - Chinese
bg - Bulgarian	en - English	hi - Hindi	ja - Japanese	pa - Punjabi	sv - Swedish
cs - Czech	es - Spanish	hr - Croatian	ko - Korean	pl - Polish	tr - Turkish
da - Danish	fi - Finnish	hu - Hungarian	lt - Lithuanian	pt - Portuguese	uk - Ukrainian
de - German	fr - French	id - Indonesian	nl - Dutch	ru - Russian	vi - Vietnamese

Sample call ¶

curl -X POST https://api.geneea.com/v3/analysis \
-H 'Authorization: user_key <YOUR USER KEY>' \
-H 'Content-Type: application/json' \
-d '{
    "id": "1",
    "text": "The trip to Innsbruck was great.",
    "analyses": ["language"]
}'

# On Windows, use \" instead of " and " instead of '

from geneeanlpclient import g3

requestBuilder = g3.Request.Builder(analyses=[g3.AnalysisType.LANGUAGE])

with g3.Client.create(userKey=<YOUR USER KEY>) as analyzer:
    result = analyzer.analyze(requestBuilder.build(
        id=str(1),
        text='The trip to Innsbruck was great.'
    ))

print(result.language.detected)

import requests

def callGeneea(input):
    url = 'https://api.geneea.com/v3/analysis'
    headers = {
        'content-type': 'application/json',
        'Authorization': 'user_key <your user key>'
    }

    return requests.post(url, json=input, headers=headers).json()

responseObj = callGeneea({
    'id': '1',
    'text': 'The trip to Innsbruck was great.',
    'analyses': ['language']
})

print(responseObj)

Priors ¶

If you know your texts can be only in certain languages, you can specify a prior – a single language or a combination of several languages. Currently, the supported priors are:

cs,de	cs,en,sk	cs,de,es,nl,pl	de,en,es,nl	es,nl,pl
cs,en	cs,es,nl	cs,en,es,nl,pl	de,en,es,pl	nl,pl
cs,es	cs,es,pl	cs,de,en,es,nl,pl	de,en,nl,pl	en,zh
cs,nl	cs,nl,pl	de,en	de,es,nl,pl
cs,pl	cs,de,en,es	de,es	en,es
cs,sk	cs,de,en,nl	de,nl	en,nl
cs,de,en	cs,de,en,pl	de,pl	en,pl
cs,de,es	cs,de,en,sk	de,en,es	en,es,nl
cs,de,nl	cs,en,es,nl	de,en,nl	en,es,pl
cs,de,pl	cs,en,es,pl	de,en,pl	en,nl,pl
cs,en,es	cs,es,nl,pl	de,es,nl	en,es,nl,pl
cs,en,nl	cs,de,en,es,nl	de,es,pl	es,nl
cs,en,pl	cs,de,en,es,pl	de,nl,pl	es,pl	EU

Use the prior exactly as written above (the same order, no spaces) and pass it via the options parameter:

curl -X POST https://api.geneea.com/v3/analysis \
-H 'Authorization: user_key <YOUR USER KEY>' \
-H 'Content-Type: application/json' \
-d '{
    "id": "1",
    "text": "The trip to Innsbruck was great.",
    "options": {"lang_prior":"en,nl"}',
    "analyses": ["language"]
}'

# On Windows, use \" instead of " and " instead of '

from geneeanlpclient import g3

requestBuilder = g3.Request.Builder(analyses=[g3.AnalysisType.LANGUAGE], customConfig={'options': {'lang_prior': 'en,nl'}})

with g3.Client.create(userKey=GENEEA_API_KEY) as analyzer:
    result = analyzer.analyze(requestBuilder.build(
        id=str(1),
        text='The trip to Innsbruck was great.'
    ))

print(result.language.detected)

import requests

def callGeneea(input):
    url = 'https://api.geneea.com/v3/analysis'
    headers = {
        'content-type': 'application/json',
        'Authorization': 'user_key <your user key>'
    }

    return requests.post(url, json=input, headers=headers).json()

responseObj = callGeneea({
    'id': '1',
    'text': 'The trip to Innsbruck was great.',
    'options': {'lang_prior':'en,nl'},
    'analyses': ['language']
})

print(responseObj)

Customization ¶

We can customize our language detection to your needs. Maybe your emails contain error messages in English, or product names sounding French, etc.