|
|
|
|
|
|
by dsyko
3539 days ago
|
|
|
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. |
|
|