Users have expressed an interest in using the framework to evaluate the transcriptions generated by speech detection components. Since the FiftyOne tool is geared towards images, it may not be a good fit for this. We should determine if other tools exist that we can leverage. Some ideas:
Users have expressed an interest in using the framework to evaluate the transcriptions generated by speech detection components. Since the FiftyOne tool is geared towards images, it may not be a good fit for this. We should determine if other tools exist that we can leverage. Some ideas:
compute-wer