List your company
ResearchFormerly WizWand

Automatic Speech Recognition on Multilingual ASR Corpus Tatar - tt (test)

6.5CERembedding-based H-Softmax
6.3567.3288.39.272Jan 29, 2025
Updated 4mo ago

Evaluation results

MethodLinks
embedding-based H-SoftmaxTraining=All, Embedding Model=XLM-XL, Distance Metric=Wtd Corr2025.016.5
SoftmaxTraining=All, Softmax Type=Standard Softmax2025.016.7
embedding-based H-SoftmaxTraining=All, Strategy=Mono-Map, Weighting=Avg CB2025.016.7
embedding-based H-SoftmaxTraining=All, Embedding Model=XLM-base, Distance Metric=2-med S-Euc2025.016.9
embedding-based H-SoftmaxTraining=All, Embedding Model=LABSE, Distance Metric=Median Euc2025.016.9
embedding-based H-SoftmaxTraining=All, Embedding Model=XLM-large, Distance Metric=2-med S-Euc2025.017
H-Softmax (Huffman)Training=All, Softmax Type=Huffman2025.017.7
embedding-based H-SoftmaxEmbedding Source=XLM-base, Clustering Method=2-medoids, Distance Metric=Standardized Euclidean2025.019
embedding-based H-SoftmaxEmbedding Source=XLM-XL, Clustering Method=Weighted, Distance Metric=Correlation2025.019.1
embedding-based H-SoftmaxEmbedding Source=Mono-Map, Clustering Method=Average, Distance Metric=City Block2025.019.2
Huffman H-SoftmaxOutput Strategy=H-Softmax, Clustering=Huffman2025.019.6
embedding-based H-SoftmaxEmbedding Source=LABSE, Clustering Method=Median, Distance Metric=Euclidean2025.019.6
embedding-based H-SoftmaxEmbedding Source=XLM-large, Clustering Method=2-medoids, Distance Metric=Standardized Euclidean2025.019.8
SoftmaxOutput Strategy=Softmax2025.0110.1