Has background noise been minimized? It is hard, and in some cases impossible, for voice recognition technology to properly function in a noisy environment. If other people are yelling, talking loudly, or otherwise making loud noises in the background then speech recognition performance can be severely degraded. If a person cannot accurately assess a speaker in a loud environment, neither can a machine.
-
- Separate recording stations as much as possible from each other.
-
- Increase the distance from noise sources and decrease the distance between the speaker and the speaker’s microphone.
- Use a headset or directional microphone, if possible.
- The ambient noise level should be similar to a study room or library, not a cafe. Nobody should be yelling or raising their voice.
Is the microphone volume set too low or is the speaker’s voice too soft? While the speaker should not talk loudly, they must also not whisper. If the microphone sensitivity is too low and the speaker is too soft, it may not be possible to hear the recorded audio. Please check your OS microphone settings and increase microphone input volume, if necessary. The average input level for a speaker’s microphone reading should fall in the middle of the allowable range.
If the issue continues please reach out to Support by selecting the ? button on the Learning Portal.