speaker independent recognition

Speaker independent recognition is speech recognition trained to be used by any speaker. They are particulatly useful for systems such as telephone help lines, where there are very large numbers of users who may each use the system for a short period of time. While the term 'speaker independent' means they are not focused on a single user, often there is an intended or unintended bias towards particular groups of people. For example, many systems are less effective for elderly people, small children or those with strong accents.

The opposite is speaker dependent recognition, which is trained for a single individual.