-
I recently found this repo and I think it's got everything I'm looking for my project. This is exciting!
I've been following a beginners tutorial (https://gist.github.com/kylebgorman/7d406f577ef192…
-
./tools/decode.sh: line 65: 24089 Aborted decoder_main --rescoring_weight $rescoring_weight --ctc_weight $ctc_weight --reverse_weight $r
everse_weight --chunk_size $chunk_size --wav_s…
-
Hi Team,
I am trying latest itn model t5 for inference.(https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/itn_en_t5)
Getting error as the following:
Error executing job with overrid…
-
我用CtcWfstBeamSearch进行解码，Partial result top 1得分是-13.8233，但是经过FinalizeSearch后，top 1得分降到了-31.7852，有时候top1因为得分降太多被其他超过了。请问这是语言模型的问题还是声学模型的问题？
https://github.com/wenet-e2e/wenet/blob/f393be9bc816fef57c6dc…
-
**Describe the bug**
Following the installation guide in README installs the latest transformers version. However the starter code provided here https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nem…
-
Thank you for open source gpu wfst decoder！
Can you please provide a C++ interface? We input the acoustic model logsoftmax, TLG.fst and word.txt, and then decode on GPU WFST directly.
Thanks
-
**Describe the bug**
I decoded several files in continuous decoding mode and did a formal WER calculation. To my surprise, the pattern that I see is that half the segments are generally good, and the…
-
GPU T4
Riva 2.10
TTS es multispeaker fastpitch-hifigan
I increased the default input when using TTS SSML tags because I need to use 1000 characters, when using it in offline mode there is a big a…
-
**Describe the bug**
when i declare some states with pynini.string_map it cant find the shortest pass of its inverse for no reason
**Steps/Code to reproduce bug**
```
import pynini
from py…
-
**Describe the bug**
I cannot use the "STT En FastConformer Hybrid Transducer-CTC Large Streaming 80ms". First of all, the NGC catalog instructions don't work. But more importantly, there is a shap…