Grammar doesn't work with trained model
Open
- Dominant language
- Jupyter Notebook
- Stars
- 15.1k
- Forks
- 1.8k
- PR merge metrics
- No merged PRs in 30d
Description
I had used mini_librispeech to train the model used in vosk, when i used the HCLr.fst to recognize the fixed text GRAM defined as "ONE TWO THREE FOUR ...", the recognition results of vosk is wrong; while i used the HCLG.fst ,the recognition results can be ok. I have no idea how to get the correct model of HCLr.fst. Looking forward to your reply！
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.