Speech Recognition On Hub500 Switchboard
Metrics
CallHome
SwitchBoard
Results
Performance results of various models on this benchmark
Model Name | CallHome | SwitchBoard | Paper Title | Repository |
|---|---|---|---|---|
| Jasper DR 10x5 | 16.2 | 7.8 | Jasper: An End-to-End Convolutional Neural Acoustic Model | |
| LAS + SpecAugment (with LM, Switchboard mild policy) | 14.6 | 6.8 | SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition | |
| LAS + SpecAugment (with LM, Switchboard strong policy) | 14 | 7.1 | SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition | |
| CTC-CRF | 18.4 | 9.7 | CAT: A CTC-CRF based ASR Toolkit Bridging the Hybrid and the End-to-end Approaches towards Data Efficiency and Low Latency | |
| Espresso | - | - | Espresso: A Fast End-to-end Neural Speech Recognition Toolkit |
0 of 5 row(s) selected.