What is the difference between Dragon NaturallySpeaking SDK Server Edition and Dragon NaturallySpeaking SDK AudioMining?
Depending on the use case, the Dragon SDK contains two sets of models that users can integrate within a server environment to run recognition: Dragon NaturallySpeaking SDK Server (DSS) Edition and Dragon SDK AudioMining. DSS is a speaker-dependent system, designed to create precisely accurate transcriptions. This model requires separation of speaker audio channels and yields significantly better results when specific user profiles are defined and refined over time. A use case for DSS could be providing transcriptions of recorded dictations of known speakers. Alternately, AudioMining is a speaker-independent system, designed to gather rough transcriptions that could be useful in indexing, word search, etc. Unlike DSS, user profiles arent necessary, enabling offline recognition from individuals or groups of speakers. The use case for AudioMining may be an organization listening to all audio to identify various content, such as harassment.
