Version 4.0.1
· 2 min read
In this release we have made several internal improvements mainly towards deployments in isolated, air-gapped environments.
Further minor improvements and fixes in configuration and administration:
- New, more readable and more visually appealing console welcome screen
- Welcome screen is now also displayed when connecting remotely via SSH
- Further improved detection of "system is ready" state during startup process
- Fixed issue when the (auto)configuration script did not properly enable some Speech To Text languages, which could make the Virtual Appliance nonfunctional
- The (auto)configuration script now also updates the number of time-slicing replicas in NVIDIA GPU plugin configuration file
Included Components
- Age Estimation 1.1.0
- Audio Manipulation Detection 1.0.0
- Audio Quality Estimation 3.62.0
- Deepfake Detection 2.2.0
- Emotion Recognition 1.2.1
- Enhanced Speech to Text Built on Whisper 1.10.0
- Gender Identification 1.4.0
- Keyword Spotting 1.1.0
- Language Identification 1.7.0
- Replay Attack Detection 1.0.0
- Speaker Diarization 1.6.0
- Speech to Text Phonexia 6th Generation 3.62.0
- Time Analysis of Speech 3.62.0
- Voice Activity Detection 1.2.0
- Voiceprint Comparison 1.4.0
- Voiceprint Extraction 1.6.0