Skip to main content
Version: 4.0.0-rc1

FAQ

Why is the transcription not precise enough?

This may be caused by several factors.

  1. The audio quality is very low, or the speech is not understandable from the recording. Please note that Speech to Text is not more precise than a human ear, which means that it should not be used for audio that is not comprehensible for the people who speak the same language as the recorded speaker.
  2. The speaker has a strong non-native accent or is a speaker of a marginal dialect that hasn’t been used in the training of this technology.
  3. There is background noise or music that deteriorates recording quality.
  4. The transcription language you chose is different from the one in the recording.
  5. The recording contains a language that is not part of Phonexia’s portfolio.

What can I do to improve the processing speed?

In case you are using Enhanced Speech to Text Built on Whisper, run the technology on GPU to improve the processing speed.