[micro_wake_word] Update documentation to discuss how voice activity is detected (#4097)

This commit is contained in:
Kevin Ahrendt 2024-07-30 22:15:41 -04:00 committed by Jesse Hills
parent 9a977067ab
commit 88d352c168
No known key found for this signature in database
GPG Key ID: BEAAE804EFD8E83A

View File

@ -55,7 +55,7 @@ Configuration variables:
- **probability_cutoff** (*Optional*, percentage): The probability cutoff for voice activity detection. - **probability_cutoff** (*Optional*, percentage): The probability cutoff for voice activity detection.
If the probability is below this value, then no wake word will be accepted. If the probability is below this value, then no wake word will be accepted.
A larger value reduces the number of false accepts but increases the number of false rejections. A larger value reduces the number of false accepts but increases the number of false rejections.
- **sliding_window_size** (*Optional*, int): The size of the sliding window for voice activity detection. The maximum of the probabilities in the sliding window is compared to ``probability_cutoff`` to determine if voice activity is detected. - **sliding_window_size** (*Optional*, int): The size of the sliding window average for voice activity detection. The average probability is compared to ``probability_cutoff`` to determine if voice activity is detected.
The ``probability_cutoff`` and ``sliding_window_size`` are provided by the JSON file but can be overridden in YAML. A default VAD model is provided with the ``vad`` configuration variables, but a different model can be overridden in YAML. The ``probability_cutoff`` and ``sliding_window_size`` are provided by the JSON file but can be overridden in YAML. A default VAD model is provided with the ``vad`` configuration variables, but a different model can be overridden in YAML.