MetricServer : Load, Predict, Analyse
Process speech file pairs through a selection of industry standard speech and audio quality metrics using MetricServer from Opale Systems. Generate Mean Opinion Score (MOS) estimates and other key metrics. The multi-user MetricServer has web-based and automation interfaces.
CONTACT US REQUEST DEMO
Targets
Mobile Operator, SIP trunk providers and Managed Service Providers to analyze speech quality metrics and end-user experience/QoE. Developers and Quality Assurance engineers working with speech and audio medias. Speech performance analysis requiring ITU-T Rec. P.863 POLQA. Postprocessing MOS and Voice Quality analysis on degraded speech files captured on any Test Systems.
How Does It Work?
Web-based UI or Python API, MetricServer has an easy-to-use workflow-oriented approach which guides users through the five-step test cycle as shown below:
Automation
Using the platform-independent Python API, MetricServer can be integrated into any proprietary test system. The reference and degraded speech files are passed to MetricServer by the third-party test application.
The results are returned to the third-party test application, to a database or stored in CSV file.
Benefits
-
Support ITU Standards algorithms and more (PESQ, POLQA®, VISQOOL, PEAQ Music analysis)
- A range of popular metrics : MOS, Signal Latency, Percentage of speech loss, Speech/Noise levels
- Opale Systems speech transmission performance expertise – over three decades
- Logical workflow, easy to use
- Automation scripts capability with platform-independent Python library
- Multi-user Web-based User Interface
- Graphical and textual presentation in browser application, for a detailed understanding of speech quality performance
- High scalability : Thousands of audio files processed per day
Voice Quality Troubleshooting
with Opale extended library
With exclusive Opale Systems advanced Speech Performance Analysis (SPA) library, enpower your analysis capability when the MOS score is poor.
View waveforms analysis
Frame by frame time offsets
Frame by frame scores
Speech Loss analysis