-
Hello, I try to create a model using `public_series_1` from Russian dataset [open_stt](https://github.com/snakers4/open_stt). I use this [recipe](https://github.com/flashlight/wav2letter/blob/master/r…
-
**Tasks**
- [x] Coding architecture of AWE model.
- [x] Setting up training and testing pipeline, dataloader + trainer
- [ ] Devising better strategy for force alignment
- [ ] Integrate ASR in f…
-
Hello, thanks for the nice implementation and pretrained model of Korean ASR.
I am testing with the code, but I couldn't find the pretrained checkpoint for this kenlm model. Could you provide the c…
-
Transducer class 코드에서.
Q1.
pred != 0 ---> 이 의미는 blank가 아닌 character를 예측한 것으로 transducer 구조(첨부한 그림 참조)에서 수직으로 이동한 것에 해당합니다.
이 경우에는 (encoder time step) t가 증가하면 안되는 것 아닌가요?
![image](https://user-ima…
-
**Describe the bug**
I am trying to use an external LLM to rescore the results of beam search from Conformer-CTC model.
When trying to get the beam search results with the `eval_beamsearch_ngram…
-
Hi every body,
I want to use external language model for ASR. according to documentation found in openseq2deq page, it must be launched in 2 steps:
1. run inference to dump logits to pickle.
2. …
-
Hi,
I would like to use the pretrained acoustic model for English but use it in combination with a new in-domain language model, for which I have to generate pronunciations.
I am used to the Kal…
-
**Describe the issue**
I am trying to install espnet toolkit. In the process I have run the following commands:-
1. git clone https://github.com/espnet/espnet
2. cd /tools
3. rm -f activate_python…
-
### A description of what we have done
1. We elaborate a **Spanish training** using this typical example [architecture](https://github.com/facebookresearch/wav2letter/blob/v0.2/tutorials/1-librispe…
-
the current UI is crap. changing this to/adding a RecognitionService would make it possible to use it from konele (https://github.com/alphacep/vosk-android-demo/issues/126#issuecomment-784579653).