Automatic Speech Recognition on Multilingual ASR Corpus Kyrgyz - ky (test)
13.9CERH-Softmax (Huffman)
Evaluation results
| Method | Links | |
|---|---|---|
| H-Softmax (Huffman)Training=All, Softmax Type=Huffman2025.01 | 13.9 | |
| embedding-based H-SoftmaxTraining=All, Embedding Model=LABSE, Distance Metric=Median Euc2025.01 | 13.9 | |
| SoftmaxTraining=All, Softmax Type=Standard Softmax2025.01 | 15.2 | |
| embedding-based H-SoftmaxTraining=All, Embedding Model=XLM-XL, Distance Metric=Wtd Corr2025.01 | 15.8 | |
| embedding-based H-SoftmaxTraining=All, Embedding Model=XLM-base, Distance Metric=2-med S-Euc2025.01 | 16 | |
| embedding-based H-SoftmaxTraining=All, Embedding Model=XLM-large, Distance Metric=2-med S-Euc2025.01 | 16.3 | |
| embedding-based H-SoftmaxTraining=All, Strategy=Mono-Map, Weighting=Avg CB2025.01 | 16.6 | |
| embedding-based H-SoftmaxEmbedding Source=LABSE, Clustering Method=Median, Distance Metric=Euclidean2025.01 | 17.3 | |
| embedding-based H-SoftmaxEmbedding Source=XLM-large, Clustering Method=2-medoids, Distance Metric=Standardized Euclidean2025.01 | 17.7 | |
| embedding-based H-SoftmaxEmbedding Source=XLM-XL, Clustering Method=Weighted, Distance Metric=Correlation2025.01 | 18.1 | |
| embedding-based H-SoftmaxEmbedding Source=Mono-Map, Clustering Method=Average, Distance Metric=City Block2025.01 | 18.2 | |
| embedding-based H-SoftmaxEmbedding Source=XLM-base, Clustering Method=2-medoids, Distance Metric=Standardized Euclidean2025.01 | 18.6 | |
| SoftmaxOutput Strategy=Softmax2025.01 | 19.8 | |
| Huffman H-SoftmaxOutput Strategy=H-Softmax, Clustering=Huffman2025.01 | 20.8 |