Skip to main content
EURASIP Journal on Audio, Speech, and Music Processing

Table 6 Comparison of real-time factor for language recognition systems

From: Homogenous ensemble phonotactic language recognition based on SVM supervector reconstruction

 

Baseline

Relative

Functional

Perturbational

  

SSR

SSR

SSR

Decoding

0.11

0.11

0.11

0.11

SV prod.

2.63×10−6

0.06

2.67×10−6

2.64×10−6

Total

0.11

0.17

0.11

0.11

  1. HU frontend, NIST LRE 2009, 30-s test. CPU: Xeon E5520@2.27 GHz, RAM: 8 GB, single thread. SV prod., super vector product.
Back to article page