-
cuda: 11.3, pytorch: 1.10.1, python: 3.8.5
I trained a conformer transducer model using aishell recipe, but got this as follows, could anyone give me a hand?
2022-01-08 19:48:09,317 (asr_train:563)…
-
Hi! Thanks for this great project! :)
I am trying to tune a Transformer RNN-T on Aishell-1 dataset, following your config in `egs/vivos/asr1/conf/tuning/transducer/train_transformer_transducer.yaml` …
-
Is there a way to compute phonetic posteriorgrams (PPG) from pretrained ASR models? It would be very helpful for tasks like voice conversion.
-
안녕하세요, joint-ctc-attention-transformer 모델을 training시키고있는데 생각보다 훈련시간을 더 늘려야할것같아서
last model checkpoint 를 사용하여 이어서 훈련을 시키고자 합니다.
config에서는 resume=True로 바꿔주었는데
argument에서 추가로 넣어주어야하는 옵션이 있을까요?
res…
-
How can I get a space(blank) it self in tokenizer vocab?
I am trying to finetune the stt_en_conformer_transducer_large model with grapheme in korean. I changed some code to load labels with chardata…
-
## ❓ Questions and Help
I'm trying to install fairseq but i can't.
Whenever I execute this code `pip3 install fairseq`, I get this overwhelmingly long error.
Collecting fairseq
Using cac…
-
1. For streaming decode, when we use **rescore decode method**, we use the encoder output + CTC to get the decode result of the current chunk, and then we use decoder to rescore the CTC decode result,…
-
The encoder outputs are fed into decoder entirely, so the encoder-decoder attention attends to the whole sequence. Right? Why not use monotonic attention?
-
Hi, did you finish the implement of Joint CTC-triggered attention decoding algorithm?
I complete the Streaming Transformer model train and finished the train with TensorFlow, but when I tried to f…
-
I am using the recipe posted here https://github.com/speechbrain/speechbrain/issues/633#issuecomment-816603992
by @TParcollet
After two days of training the WER is still around 55-60, while the p…