New processing options for ASR related to performance and accuracy
See the following page for more detail - https://support.intelligentvoice.com/hc/en-us/articles/23793291609107-Processing-Options-For-NASRv5-Language-Models
Diarization
Diarization algorithm speaker improvements
Duplicate speaker and unknown speaker fixes
JumpToWeb
Fix for missing tag cloud and hypertree
Support for LDAP authentication
Bug Fix: Tagger processing option re-enabled
In the 6.1 release, the default processing option for tags (topics) was changed to disabled. This has been reverted.
This fixes the reported issue of missing tags for API users where the option was not specified.
This means the default behaviour will revert to creating a tagger job for all items. This requires the tagger_worker to be installed and running for the item to be marked as complete.
Bug Fix: Format conversion errors now marks items as complete with error
Previously job items that had the format conversion error were never marked as complete, this is now caught at the first stage through IV processing and appropriately marked the item as complete with an error.
If you are looking for a demo of current or new functionality, please contact your usual sales representative or sales@intelligentvoice.com.
Existing users can download the new release from their support portal, but please feel free to get in touch with your support representative to talk more about deployment options, or contact support@intelligentvoice.com.