Automatic Speech Recognition on Turkic Global concatenated (test)
14.6CERembedding-based H-Softmax
Evaluation results
| Method | Links | |
|---|---|---|
| embedding-based H-SoftmaxEmbedding Source=LABSE, Clustering Method=Median, Distance Metric=Euclidean2025.01 | 14.6 | |
| embedding-based H-SoftmaxEmbedding Source=XLM-XL, Clustering Method=Weighted, Distance Metric=Correlation2025.01 | 14.7 | |
| embedding-based H-SoftmaxEmbedding Source=XLM-base, Clustering Method=2-medoids, Distance Metric=Standardized Euclidean2025.01 | 14.8 | |
| embedding-based H-SoftmaxEmbedding Source=Mono-Map, Clustering Method=Average, Distance Metric=City Block2025.01 | 14.8 | |
| Huffman H-SoftmaxOutput Strategy=H-Softmax, Clustering=Huffman2025.01 | 14.9 | |
| embedding-based H-SoftmaxEmbedding Source=XLM-large, Clustering Method=2-medoids, Distance Metric=Standardized Euclidean2025.01 | 15.1 | |
| SoftmaxOutput Strategy=Softmax2025.01 | 16.6 |