In some embodiments, the transcription system 130 may be configured to obtain transcriptions using different transcription generation techniques. The transcription generation techniques may include the use of different speech recognition systems. For example, a first transcription generation technique may include the use of the first speech recognition system 134 and a second transcription generation technique may include the use of the second speech recognition system 136.

In some embodiments, a particular transcription generation technique may initially be selected to obtain transcriptions. In these or other embodiments, the particular transcription generation technique may initially be selected based on one or more factors such as cost, accuracy, user preference, etc. Additionally or alternatively, some factors may be weighted more than others in determining which transcription generation technique to initially select. For example, cost may be weighted higher than accuracy in some instances.

In some embodiments, after the initial selection of the particular transcription generation technique, another transcription generation technique may be selected to obtain transcriptions instead of the particular transcription generation technique. In these or other embodiments, the selection of the other transcription generation technique may be based on user ratings of transcriptions