-
Hi,
My question is about data prep part. I notice that Nemo has separate scripts for different dataset, e.g. get_librispeech_data.py, process_aishell_data.py. Also, you have scp_to_manifist.py whic…
-
My dataset contains shorter utterances and for the speaker verification task(low-resource language), I had split my test utterances for 1sec and 2sec.
Can you please help me in providing a feasible …
-
Hi, thanks for creating this guide, it really useful for me who don't understand much about bash scripting. I managed to create the model with wav and txt file only.
Can you create a guide for fin…
-
I want re-procedure result like when i use compute log-filterbank energy of lib: python_speech_feature by using torchaudio.
this is my code:
```
# load audio data by librosa
path_audio = "audio_…
-
After compiling kaldi as mentioned in the instructions, while running egs/vcc20/vc1_task1/run.sh, I encountered an error on stage 3 saying compute-vad command not found.
```
[x] python=3.8.8 (defa…
-
Hi,
I wonder if there is any way to put checkpoints during TDNN kaldi training,
so that , if somehow an error occurs at some point, we can rollback to a safe checkpoint and continue from there on,
…
-
How do i get a recompiled rnnlm rescoring files when updating my vocab, and do i need it?
-
We are struggling to build a noise robust speech recognition model.
I have questions about the noise robust speech recognition model using Espnet.
(1) Which method do you think is better on buildi…
ken57 updated
3 years ago
-
I want to decode through VTLN model and i-vecotr model.
How do I achieve it?
I don't see any "decode.sh" that has vtln+ivecotr.
thanks.
-