Training on GPU taking very long time

[Abraham Nyongesa]

Hello members. I am training diarization model under libri_css on my GPU enabled machine GeForce RTX 2080 and my machine has 32GB RAM but the script is taking forever to train.

train_dnn_raw.py for 3 epochs and 38 iterations is taking forever. For two days it has done only 2 iterations. How can I solve the problem so that it speeds up?

[Desh Raj]

Training an x-vector extractor on a single GPU can take a long time, unfortunately. I would suggest using the pretrained model from here: https://kaldi-asr.org/models/12/0012_diarization_v1.tar.gz.

Desh

On Wed, Mar 1, 2023 at 11:09 AM Abraham Nyongesa <abraham.n...@gmail.com> wrote:

Hello members. I am training diarization model under libri_css on my GPU enabled machine GeForce RTX 2080 and my machine has 32GB RAM but the script is taking forever to train.

train_dnn_raw.py for 3 epochs and 38 iterations is taking forever. For two days it has done only 2 iterations. How can I solve the problem so that it speeds up?

--
visit http://kaldi-asr.org/forums.html to find out how to join.
---
You received this message because you are subscribed to the Google Groups "kaldi-developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kaldi-develope...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/kaldi-developers/CAHXyX0n1OdSkvb8aaa6d5BCF5DhAk%3Dv-t4CmX2HDQs4cv7Sf8g%40mail.gmail.com.

[Jayenthiran Pukuraj]

Hi Desh,

I've been learning and working with Kaldi for the past two months, and I'm happy to report that I’ve successfully managed to decode my own audio using the LibriSpeech dataset. However, my requirement is specifically for British English (Received Pronunciation - RP), while LibriSpeech is American English.  

  Could anyone kindly guide me on the best  dataset for British english?

even its external dataset also fine we can able to train in kaldi is happy 

please help me on this