At Intento, we run multiple evaluations for our clients for various language pairs and domains, giving us both a bird’s-eye view of this vast landscape and in-depth knowledge of each provider. Benefit from our expertise by downloading the current MT landscape!
Pre-educated models based on generic data without a specific domain, meaning that these models are not pre-adjusted to one particular industry or specialization, such as Legal or Medical translations.
Follows the same logic as Generic Stock Models, in that users do not customize the MT models in any way. However, they do fit under a specific domain, relying on the context surrounding a particular industry.
Allows users to customize the MT models by applying their own glossaries. Depending on the provider, terminology can be used while training custom models or for adjusting machine translation results.
Combines custom models produced by users who have added their text to the original stock model with the pre-educated stock models based on generic data, allowing users to personalize their models.
Puts customization in the provider’s hands rather than providing their own glossaries. The user comes directly to the provider and requests a customized model directed towards a particular domain.