We provide two types of pretrained model in WeNet to facilitate users with different requirements.
-
Checkpoint Model, with suffix .pt, the model trained and saved as checkpoint by WeNet python code, you can reproduce our published result with it, or you can use it as checkpoint to continue.
-
Runtime Model, with suffix .zip, you can directly use
runtime modelin our x86 or android runtime, theruntime modelis export by Pytorch JIT on thecheckpoint model. And the runtime models has been quantized to reduce the model size and network traffic.
The pretrained model in WeNet follows the license of it's corresponding dataset.
For example, the pretrained model on LibriSpeech follows CC BY 4.0, since it is used as license of the LibriSpeech dataset, see http://openslr.org/12/.
Here is a list of the pretrained models on different datasets. The model structure, model size, and download link are given.
| Datasets | Languages | Checkpoint Model | Runtime Model | Contributor |
|---|---|---|---|---|
| aishell | CN | Conformer | Conformer |  |
| aishell2 | CN | Conformer | Conformer | |
| gigaspeech | EN | Conformer | Conformer | |
| librispeech | EN | Conformer | Conformer | |
| multi_cn | CN | Conformer | Conformer |  |
| wenetspeech | CN | Conformer | Conformer |  |