Skip to main content

Language

Note: Multi-language support is an experimental feature. If you have any issues with a particular language or need a language that isn't supported, please let us know.

Language

Machined supports writing articles in multiple languages using OpenAI's language models. Language is set when you create a cluster and define your audience.

Supported languages​

Machined officially supports 58 languages through OpenAI's models. Many additional languages work unofficially, though quality may vary.

English generally produces the best results, as it has the most training data and refinement in the underlying AI models.

How language works​

When you create a cluster, you specify the language as part of your audience definition. This affects:

  • Article content: All articles in the cluster are written in the specified language
  • Keywords and titles: Generated keywords and article titles are in the specified language
  • Research: The research process focuses on sources in the target language

The language setting does not change the Machined interface, which remains in English.

Language quality​

All officially supported languages should work similarly, though AI models can be unpredictable. Quality depends on:

  • The amount of training data available for that language
  • The complexity of your topic
  • The availability of research sources in that language

English typically produces the most consistent results due to more extensive training data.

Voices and tones in different languages​

All built-in voices and tones should work across all supported languages, though the exact style and nuance may vary depending on the language. Some tones and styles translate more naturally into certain languages than others.

Setting language​

Language is specified when you create a cluster as part of your audience definition. All articles in that cluster will use the same language. If you need content in multiple languages, create separate clusters for each language.