Characterizing accuracy of ensemble models for automatic speech recognition

A non-transitory computer readable medium storing instructions that, when executed by a processor, cause the processor to perform the steps of:

• storing per-character differences between a first set of characters and a second set of characters in a matrix with a fixed width;
• encoding the per-character differences in the matrix into a vector of the fixed width;
• and
• outputting the vector as a representation of a pairwise error rate between the first set of characters and the second set of characters.

A system, comprising:

• a memory that stores instructions;
• and
• a processor that is coupled to the memory and, when executing the instructions, is configured to:
• store per-character differences between a first set of characters from a first transcription of the recording and a second set of characters from a second transcription of the recording in a matrix with a fixed width;
• encode the per-character differences in the matrix into a vector of the fixed width;
• and
• output the vector as a representation of a pairwise error rate between the first transcription and the second transcription.