Minimum amount of audio data processed by the endpointer.
|
Value |
Integer: 1–1999. |
|
Default |
800 (bytes) |
|
How to set |
Shown in precedence order (see Rules of parameter precedence): 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 reduce latency. Useful when sending numerous, small buffers of audio data. |
|
Performance |
Because each buffer incurs a fixed amount of CPU overhead, it is more efficient to process several buffers at the same time when the buffers are very small. When setting this parameter, you must balance CPU load and system response time. Lesser values increase the amount of CPU cycles while improving system response (and reduce caller-perceived delays). Higher values reduce CPU load at the cost of system response. |
This parameter defines a minimum amount of endpointer processing: when a buffer is smaller than this value, the endpointer waits for additional requests to the endpointer and then processes the buffers in one sequence. (Until the buffers are processed, it does not change its previous endpointer state or status return values.)
