Some automatic translation vocabulary

A number of technical terms are used in this chapter. The AI Speak page has some details and the history of natural language processing.

From a broader perspective, automatic translation tools are accessible online and can be used in many ways:

• Directly, by copy-pasting bits of text in one language and obtaining the same text in another language;
• By submitting to the website files in some predefined formats. The whole file will be translated, preserving the structure and formatting of the file;
• Through APIs (an application programming interface is a piece of code programmers can use in their software to facilitate translation tools).

Automatic transcription also makes use of artificial intelligence. This consists of transforming a voice input into a text output. It can be done either online or on offline videos or audio recordings. Some video conference platforms can obtain subtitles which can then be used to improve accessibility and/or to understand speech from another language.

Voice-synthesis techniques take text and a voice model to speak the text using that voice. Voice model can be standard or can also be trained to correspond to real people.

Text-generation tools are used to generate new text through artificial intelligence. This new text can be based on existing text, such as abstracts, simplifications or reformulations of existing text, or based on conversational models, where AI will be prompted on a topic.

These techniques can be used separately or be combined, in order to propose seamless multilingual conversations.