With Trainer, you can train custom machine translation models on different platforms from the same interface.
Important! To use Trainer, you need to have a direct contract with the MT provider.
How MT customization works
When you train a custom MT model, you customize a provider's existing MT model to make it translate specific content in a specific way. To customize a model, you need to give it parallel text segments: source texts and translations into a target language. The model processes the segments and learns to translate words and phrases in a way that you want it to.
Compared to stock MT models, custom models produce translations that are closer to reference translations.
Get connected accounts for all providers
To use Trainer, start accounts with all MT providers whose models you want to customize. Then connect those accounts to your Intento account.
To connect a Google account, follow the instructions here.
To connect a ModernMT account, follow the instructions here.
Training a ModernMT model is free.
The price of training a Google model depends on the amount of training data. The minimum is approximately $100 and the maximum per training is fixed at $300.
To train custom models, you need one or more .tmx files with parallel text segments. For best training results, the total number of segments should be between 15,000 and 100,000. Each file must not be greater than 250 MB.
Some MT providers need a minimum number of segments to train a custom model. Google needs at least 1,000 segments. ModernMT has no limitations.
Start working with Trainer
Go to Intento MT Studio and select Trainer.
If this is the first time you work with Trainer, select Create training project to begin.
If this isn't your first project, you'll see the list of your old projects. Select Start a new training to create a new project.
Upload your .tmx files. When you upload files, Trainer processes the data to remove duplicated segments. Processing takes a few minutes.
When your files are processed, enter a project name in the Name field. Choose the source and target languages in the From and To fields.
Choose a provider in the Provider field.
Choose an account in the Connected account field. If you're having trouble with accounts, contact your organization's admin or Intento support.
To train a model from another provider, select Add provider and choose another provider and connected account.
Select Start training.
Wait for results
A ModernMT custom model takes a few seconds to train. When the model is ready, Trainer shows you the training results.
A Google model takes several hours to train. You can come back to Trainer later and see the status of your models on the main page: Completed if the model is ready or In progress if it's still training. If training has failed, you'll see Error and a redo arrow — select the arrow to start training again. Select the provider name to see the training results.
The results page shows the model's status and the Model ID and Model key — they are needed to translate with this model.