From 95f888fd6a30f6d2fc5614347522eb854dcffbd6 Mon Sep 17 00:00:00 2001 From: Patrick von Platen Date: Fri, 24 Sep 2021 09:53:37 +0200 Subject: [PATCH] Update README.md --- examples/pytorch/speech-recognition/README.md | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/examples/pytorch/speech-recognition/README.md b/examples/pytorch/speech-recognition/README.md index 53a1c2863d..43832655ed 100644 --- a/examples/pytorch/speech-recognition/README.md +++ b/examples/pytorch/speech-recognition/README.md @@ -30,15 +30,14 @@ In the script [`run_speech_recognition_ctc`], we first create a vocabulary from --- **NOTE** -It is currently not recommended to make use of `--preprocessing_num_workers`. -If however, you wish to use multi-processing for data preprocessing by setting `--preprocessing_num_workers` > 1, +If you wish to use multi-processing for data preprocessing by setting `--preprocessing_num_workers` > 1, please make sure to set the environment variable `OMP_NUM_THREADS` to 1 as follows: ```bash OMP_NUM_THREADS=1 python run_speech_recognition_ctc ... ``` -If the environment variable is not set, the training script might hang, *i.e.* see: https://github.com/pytorch/audio/issues/1021#issuecomment-726915239 +If the environment variable is not set, the training script might freeze, *i.e.* see: https://github.com/pytorch/audio/issues/1021#issuecomment-726915239 ---