News for hackers. Stuff that matters.
Re:Deepgram – Find Damning Soundbites by dsyko on Monday October 10, 2016 @02:45AM (#12675467)
20% is very low. We use machine transcription at work, and although the per-word confidence of machine algorithms varies wildly given the quality of the audio and the speaker's mannerisms. It can easily get into the 80% range with good audio, especially televised audio where people are close to microphones and there is not a lot of non-speech noise.

There are no comments on this comment yet.