Armenian
Emerging2papers using it
2024first seen
The 'Armenian' dataset/benchmark contains speech data used to evaluate the effectiveness of a novel pseudo-labeling algorithm in improving automatic speech recognition performance in low-resource language settings.