Release notes 3.4.0
Release date: 6th October 2022
Summary
This page highlights all the changes, new features and bugs addressed within the LumenVox Containers version 3.4.0 release. This change affects both Speech and Voice biometric products.
Although not tied to a version number a simplified Kubernetes installation process utilizing KubeAdmin has been made available
Highlights
Further improvement of MRCP connection interface into speech products
Improvements have been made to overall speech product stability and scalability
Playback detection enhancements implemented for voice biometrics
Liveness detection anti-spoofing measure implemented for voice biometrics
Voice biometric enhancements to voice quality errors
Licensing counters & reporting implemented for all products
Bug fixes identified in version 3.3.0 are addressed
What’s new on LumenVox Cloud 3.4.0
New Features
Liveness detection anti-spoofing measure implemented for voice biometrics
Updates
Improvements have been made to the following:
General stability & scalability of voice biometric & speech products
MRCP connection interface
Playback detection performance and functionality has been improved
Voice biometric enhancements to voice quality errors
Licensing counters & licensing reporting implemented across all products
Bug fixes
Various including issues with the following:
CPA tests fail when unknown silence using default settings
Using an invalid abnf grammar file returns an xml file
Diagnostics test for MongoDB checks if there is a "local" DB present. Not all MongoDB deployments include a "Local" DB. Test changed to just test the connection string.
ASR indicates unhealthy during startup
Improved error message for Mongo Connection String issues that were unclear
Playback - hashes saved for enrollments are also containing the hash for a file judged as WrongContent
TTS does not react to malformed SSML
German asr logs written on asr_logs_fr
Installation notes
The following helm chart can be used
Note that for MRCP there is no helm chart but a docker compose file. MRCP will run on its own Docker virtual machine which will integrate into the Kubernetes cluster.
Upgrade procedures
Upgrade or migration from versions 1.x ,version 2.x are not supported. Version 3.4.0 must be a fresh install. Version 3.3.0 can be upgraded to version 3.4.0 by updating the protobuf files, all existing grammar cache files must be deleted prior to upgrade.
Updated API guide
Speech: LumenVox Speech API
Voice biometrics: LumenVox API Documentation
Further information can be obtained here:
Model versions as part of the release
ASR - 2.6.1
TTS - 1.0 sample rate 22
VB - 2.1.13
VB incorporates Selene 2.4.3 which was integrated into the Container stack