If an engine needs to be tuned, or trained, for a specific user then it is said to be speaker dependent. An engine that can recognise speech from many users without each user having to undergo a training regime, is classified as speaker independent. It is common for a commercial tool to be shipped with a default user model in order to provide some level of speaker independence, but be adaptable enough to provide speaker dependent operation after training. Speaker dependent operation is often needed in order to achieve acceptable recognition rates due to the great variety in the way people speak.