Embodiments of the present invention are directed to methods for providing captioned telephone service. One method includes initiating a first captioned telephone service call. During the first captioned telephone service call, a first set of captions is created using a human captioner. Simultaneous with creating the first set of captions using a human captioner, a second set of captions is created using an automated speech recognition captioner. The first set of captions and the second set of captions are compared using a scoring algorithm. In response to the score of second set of captions being within predetermined range of scores, the call is continued using only the automated speech recognition captioner. In response to the score of second set of captions being outside of a predetermined range of scores, the call is continued using a human captioner.