Google Translate

What is Googol Translate?

This is a full stack web application which imitates the lesser-known competitor application, Google Translate. You can use the API endpoint to generate your own translations.

The translation results are deterministic, like a real translation dictionary. The same inputs will result in the same outputs. Non-deterministic. See ¹.

Note: For purposes of operating with HTMX, the response is plain text, with formatted HTML hypermedia on errors.

Try it yourself

Website: googoltranslate.com

API endpoint: api.googoltranslate.com/translate

Example usage

Example JSON payload

{
    "text": "We are just an advanced breed of monkeys on a minor planet of a very average star. But we can understand the Universe. That makes us something very special.",
    "voice": "Caveman"
}

With cURL

foo@bar:~$ curl -X POST -d \
'{
    "text": "We are just an advanced breed of monkeys on a minor planet of a very average star. But we can understand the Universe. That makes us something very special.",
    "voice": "Caveman"
}' \
-X POST "https://api.googoltranslate.com/translator"

We advanced monkey type on small rock of normal fire ball. But we know big everything. That make us very special

Funny results

{
    "text": "Okay",
    "voice": "Long and convoluted"
}

In a manner that is acceptable and agreeable to the current circumstances, I hereby acknowledge your statement or request and will proceed accordingly with the understanding that my actions or responses are in alignment with your expectations.

{
    "text": "Saturn",
    "voice": "Caveman"
}

Big round rock in sky with rings

Technologies used

Frontend: HTMX, HTML, Sass

Backend: Rust, Cargo Lambda,

Deployment: AWS, Lambda, API Gateway, EC2, Route53, ACM, GitHub Workflows

APIS: OpenAI

Footnotes

1. Although the parameters for temperature and top_p have been set to 0.0, the model must still predict the translation probabilistically rather than deterministically. Deterministic outputs would have been ideal for this application, however we cannot obtain this at the moment. Perhaps in the future, there could be a truly deterministic prediction model. ↩