Docs

Read more about risk prediction and its use cases.

Getting started

To get started with ModelFront:

  1. Sign up and open the console

  2. Upload a dataset or try one of our sample datasets

  3. Hit our REST API

Console

The ModelFront console is a quick and effective way to try translation risk prediction and understand your own translation risks by graphing, slicing and monitoring ModelFront's aggregate predictions.

API

The risk API provides real-time risk predictions on each machine translation. You get a risk prediction score between 0.0 and 1.0 on each machine translation, typically a sentence or a phrase. You set your own quality threshold to you control the balance of quality, scale and speed.

Error types

ModelFront is built to catch many kinds of errors, both the common ones and the rare but painful ones.

Named entities

Errors translating names of persons, places, organisation and products

Negation

Revered meaning due to missing or inserted negation

Offensive

Offensive or ambiguous translations

Identity

Untranslated words or phrases

Noise

Garbage input or output characters, words or phrases

More

Many other types of translation errors, for example translating from the wrong language

Have another type of error you want to catch? Let us know!

Languages

ModelFront supports any language pair supported by the Google Translate API, the Yandex Translate API, the Microsoft Translator API, the IBM Watson Translator API, the Baidu Translate API or the DeepL API.

Afrikaans, Albanian, Amharic, Arabic (standard and Levantine), Armenian, Azerbaijani, Bashkir, Basque, Belarusian, Bengali (Bangla), Bosnian, Bulgarian, Burmese, Cantonese (traditional and simplified), Catalan, Cebuano, Chichewa, Chinese (traditional, simplified and classical), Corsican, Croatian, Czech, Danish, Dutch, English, Esperanto, Estonian, Fijian, Filipino (Tagalog), Finnish, French, Frisian, Galician, Georgian, German, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi, Hmong, Hungarian, Icelandic, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Kannada, Kazakh, Khmer, Klingon, Korean, Kurdish (Kurmanji), Kyrgyz, Lao (Laotian), Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Maori, Marathi, Mari (Eastern and Western), Mayan (Yucatec Maya), Mongolian, Nepali, Norwegian (Bokmål), Otomi (Querétaro), Papiamento, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Samoan, Scots Gaelic, Serbian, Sesotho, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Spanish, Sundanese, Swahili, Swedish, Tahitian, Tajik, Tamil, Tatar, Telugu, Thai, Tongan, Turkish, Udmurt, Ukrainian, Urdu, Uzbek, Vietnamese, Welsh, Xhosa, Yiddish, Yoruba, Zulu

Engines

ModelFront supports any translation API. It is trained and tested against all the major ones - Google Translate, DeepL, Microsoft Translator, Yandex.Translate (Яндекс), IBM Watson Language Translator, Baidu Fanyi (百度), Systran Translation, KantanMT, eBay Translation and Amazon Translate.

It also works with custom-trained models like those from Google AutoML and Microsoft Custom Translator, which are key to many of the applications successfully using machine translation at scale.

Customizing

ModelFront automatically improves as you use it. You can also add your own monolingual data, parallel data or labelled data to accelerate the rate at which ModelFront improves on your data.

Plans

ModelFront risk prediction API calls are charged by the character. There is a discount for prepaid calls.

The console, uploading data is free.

We also offer on-premise deployable models for very highly-scaled clients.

Support

If you have an issue or question, please [email protected].