When to update recognition models with learned statistics.
|
Value |
24-hour clock time using the hhmm format. |
|
Default |
0000 (midnight) |
|
How to set |
In Management Station set on the Nuance recognition service. If not using Management Station, set in a Recognizer configuration file (User-nrsxx.xml). |
|
Usage |
Occasionally used to perform an update at a relatively low-demand time. |
|
Performance |
Avoids caller-perceived delays by scheduling updates when applications are not busy. |
This feature is part of Recognizer’s self-learning technology. To avoid caller-perceived delays, perform updates when the recognition machine is typically not busy.
Specify the time with the hhmm format as follows:
- The time uses a 24-hour clock and hh must be between 00 and 23
- Minutes (mm) must be between 00 and 59.
- As a special case, 2400 is equivalent to 0000 (midnight).
- Leading zeros are required. For example, 0109 (1:09am) and not 109 (invalid).
Recognizer does not begin the update at the exact time specified. Instead, it begins after the next recognition event, and only if there is enough data in the adaptation statistics file.
If the Recognizer is stopped and restarted, some statistical data collected since the previous update of the models may be lost. This loss of data does not reduce recognition accuracy or corrupt the system in any way.
For an overview of adaptation, see Controlling the self-learning recognizer.
Related parameters:
