Nadira PoveyK2fsa: Training Advice for Beginners[Blog was created from our youtube interview with Daniel PoveyJan 24Jan 24
Nadira PoveyK2fsa: TensorBoard, Troubleshooting.[Blog was created from our youtube interview with Daniel Povey]Jan 9Jan 9
Nadira Poveyk2fsa-icefall: Train STT model using LibriSpeech dataset.The goal of this tutorial is to be able to train a LibriSpeech model and to evaluate our new trained model.Jan 6Jan 6
Nadira PoveyLhotse: ValueError: The number of declared samples in the recording diverged [K2-fsa/icefall]When running prepare.sh for the icefall Librispeech recipe, Stage 3 caused me the following error:Jan 2Jan 2
Nadira PoveyList of all Next-gen Kaldi tutorials from my youtube channel.64. Nadira: Train RNNLM and 2-gram for LODR decoding video [NEXT-GEN-KALDI]Dec 7, 2023Dec 7, 2023
Nadira Poveyk2fsa-icefall: How to prepare transcript_tokens.txt and 2gram.fst.txt for LODR decodingYoutube tutorial can be found here.Sep 9, 2023Sep 9, 2023
Nadira Poveyk2fsa: Create Icefall Environment for AWS Deep Learning ImageIt normally takes me approximately 10 mins to set up icefall environment. First time I did it, it took me a week to make everything work…Sep 5, 2023Sep 5, 2023
Nadira Poveyk2fsa-icefall: Building RNNLM for LODR decodingCheck tutorials here to better understand the problem that we want to solve. Youtube tutorial can be found here.Sep 5, 2023Sep 5, 2023
Nadira Poveyffmpeg: Compress audio filesWhen creating a dataset for speech processing we usually run into a problem that our datasets need 1TB or more space on our machines…Aug 23, 20231Aug 23, 20231
Nadira Poveyk2fsa:text_search ModuleNotFoundError: No module named ‘textsearch’The error above was fixed by doing pip3 install fasttextsearch.Aug 16, 2023Aug 16, 2023